diff --git a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextAnnotation.java b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextAnnotation.java
index 409b0c1adbc4..a07e16e29c2b 100644
--- a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextAnnotation.java
+++ b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextAnnotation.java
@@ -23,8 +23,6 @@ private TextAnnotation(com.google.protobuf.GeneratedMessageV3.Builder<?> builder
   }
   private TextAnnotation() {
     text_ = "";
-    confidence_ = 0F;
-    frames_ = java.util.Collections.emptyList();
     segments_ = java.util.Collections.emptyList();
   }
 
@@ -58,24 +56,10 @@ private TextAnnotation(
             text_ = s;
             break;
           }
-          case 21: {
-
-            confidence_ = input.readFloat();
-            break;
-          }
-          case 26: {
-            if (!((mutable_bitField0_ & 0x00000004) == 0x00000004)) {
-              frames_ = new java.util.ArrayList<com.google.cloud.videointelligence.v1p2beta1.TextFrame>();
-              mutable_bitField0_ |= 0x00000004;
-            }
-            frames_.add(
-                input.readMessage(com.google.cloud.videointelligence.v1p2beta1.TextFrame.parser(), extensionRegistry));
-            break;
-          }
-          case 34: {
-            if (!((mutable_bitField0_ & 0x00000008) == 0x00000008)) {
+          case 18: {
+            if (!((mutable_bitField0_ & 0x00000002) == 0x00000002)) {
               segments_ = new java.util.ArrayList<com.google.cloud.videointelligence.v1p2beta1.TextSegment>();
-              mutable_bitField0_ |= 0x00000008;
+              mutable_bitField0_ |= 0x00000002;
             }
             segments_.add(
                 input.readMessage(com.google.cloud.videointelligence.v1p2beta1.TextSegment.parser(), extensionRegistry));
@@ -96,10 +80,7 @@ private TextAnnotation(
       throw new com.google.protobuf.InvalidProtocolBufferException(
           e).setUnfinishedMessage(this);
     } finally {
-      if (((mutable_bitField0_ & 0x00000004) == 0x00000004)) {
-        frames_ = java.util.Collections.unmodifiableList(frames_);
-      }
-      if (((mutable_bitField0_ & 0x00000008) == 0x00000008)) {
+      if (((mutable_bitField0_ & 0x00000002) == 0x00000002)) {
         segments_ = java.util.Collections.unmodifiableList(segments_);
       }
       this.unknownFields = unknownFields.build();
@@ -162,83 +143,14 @@ public java.lang.String getText() {
     }
   }
 
-  public static final int CONFIDENCE_FIELD_NUMBER = 2;
-  private float confidence_;
-  /**
-   * <pre>
-   * Confidence for the track of detected text. It is calculated as the highest
-   * over all frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>float confidence = 2;</code>
-   */
-  public float getConfidence() {
-    return confidence_;
-  }
-
-  public static final int FRAMES_FIELD_NUMBER = 3;
-  private java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> frames_;
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> getFramesList() {
-    return frames_;
-  }
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  public java.util.List<? extends com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> 
-      getFramesOrBuilderList() {
-    return frames_;
-  }
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  public int getFramesCount() {
-    return frames_.size();
-  }
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  public com.google.cloud.videointelligence.v1p2beta1.TextFrame getFrames(int index) {
-    return frames_.get(index);
-  }
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  public com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuilder(
-      int index) {
-    return frames_.get(index);
-  }
-
-  public static final int SEGMENTS_FIELD_NUMBER = 4;
+  public static final int SEGMENTS_FIELD_NUMBER = 2;
   private java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextSegment> segments_;
   /**
    * <pre>
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextSegment> getSegmentsList() {
     return segments_;
@@ -248,7 +160,7 @@ public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextSegment>
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   public java.util.List<? extends com.google.cloud.videointelligence.v1p2beta1.TextSegmentOrBuilder> 
       getSegmentsOrBuilderList() {
@@ -259,7 +171,7 @@ public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextSegment>
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   public int getSegmentsCount() {
     return segments_.size();
@@ -269,7 +181,7 @@ public int getSegmentsCount() {
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   public com.google.cloud.videointelligence.v1p2beta1.TextSegment getSegments(int index) {
     return segments_.get(index);
@@ -279,7 +191,7 @@ public com.google.cloud.videointelligence.v1p2beta1.TextSegment getSegments(int
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   public com.google.cloud.videointelligence.v1p2beta1.TextSegmentOrBuilder getSegmentsOrBuilder(
       int index) {
@@ -303,14 +215,8 @@ public void writeTo(com.google.protobuf.CodedOutputStream output)
     if (!getTextBytes().isEmpty()) {
       com.google.protobuf.GeneratedMessageV3.writeString(output, 1, text_);
     }
-    if (confidence_ != 0F) {
-      output.writeFloat(2, confidence_);
-    }
-    for (int i = 0; i < frames_.size(); i++) {
-      output.writeMessage(3, frames_.get(i));
-    }
     for (int i = 0; i < segments_.size(); i++) {
-      output.writeMessage(4, segments_.get(i));
+      output.writeMessage(2, segments_.get(i));
     }
     unknownFields.writeTo(output);
   }
@@ -324,17 +230,9 @@ public int getSerializedSize() {
     if (!getTextBytes().isEmpty()) {
       size += com.google.protobuf.GeneratedMessageV3.computeStringSize(1, text_);
     }
-    if (confidence_ != 0F) {
-      size += com.google.protobuf.CodedOutputStream
-        .computeFloatSize(2, confidence_);
-    }
-    for (int i = 0; i < frames_.size(); i++) {
-      size += com.google.protobuf.CodedOutputStream
-        .computeMessageSize(3, frames_.get(i));
-    }
     for (int i = 0; i < segments_.size(); i++) {
       size += com.google.protobuf.CodedOutputStream
-        .computeMessageSize(4, segments_.get(i));
+        .computeMessageSize(2, segments_.get(i));
     }
     size += unknownFields.getSerializedSize();
     memoizedSize = size;
@@ -354,12 +252,6 @@ public boolean equals(final java.lang.Object obj) {
     boolean result = true;
     result = result && getText()
         .equals(other.getText());
-    result = result && (
-        java.lang.Float.floatToIntBits(getConfidence())
-        == java.lang.Float.floatToIntBits(
-            other.getConfidence()));
-    result = result && getFramesList()
-        .equals(other.getFramesList());
     result = result && getSegmentsList()
         .equals(other.getSegmentsList());
     result = result && unknownFields.equals(other.unknownFields);
@@ -375,13 +267,6 @@ public int hashCode() {
     hash = (19 * hash) + getDescriptor().hashCode();
     hash = (37 * hash) + TEXT_FIELD_NUMBER;
     hash = (53 * hash) + getText().hashCode();
-    hash = (37 * hash) + CONFIDENCE_FIELD_NUMBER;
-    hash = (53 * hash) + java.lang.Float.floatToIntBits(
-        getConfidence());
-    if (getFramesCount() > 0) {
-      hash = (37 * hash) + FRAMES_FIELD_NUMBER;
-      hash = (53 * hash) + getFramesList().hashCode();
-    }
     if (getSegmentsCount() > 0) {
       hash = (37 * hash) + SEGMENTS_FIELD_NUMBER;
       hash = (53 * hash) + getSegmentsList().hashCode();
@@ -520,7 +405,6 @@ private Builder(
     private void maybeForceBuilderInitialization() {
       if (com.google.protobuf.GeneratedMessageV3
               .alwaysUseFieldBuilders) {
-        getFramesFieldBuilder();
         getSegmentsFieldBuilder();
       }
     }
@@ -529,17 +413,9 @@ public Builder clear() {
       super.clear();
       text_ = "";
 
-      confidence_ = 0F;
-
-      if (framesBuilder_ == null) {
-        frames_ = java.util.Collections.emptyList();
-        bitField0_ = (bitField0_ & ~0x00000004);
-      } else {
-        framesBuilder_.clear();
-      }
       if (segmentsBuilder_ == null) {
         segments_ = java.util.Collections.emptyList();
-        bitField0_ = (bitField0_ & ~0x00000008);
+        bitField0_ = (bitField0_ & ~0x00000002);
       } else {
         segmentsBuilder_.clear();
       }
@@ -572,20 +448,10 @@ public com.google.cloud.videointelligence.v1p2beta1.TextAnnotation buildPartial(
       int from_bitField0_ = bitField0_;
       int to_bitField0_ = 0;
       result.text_ = text_;
-      result.confidence_ = confidence_;
-      if (framesBuilder_ == null) {
-        if (((bitField0_ & 0x00000004) == 0x00000004)) {
-          frames_ = java.util.Collections.unmodifiableList(frames_);
-          bitField0_ = (bitField0_ & ~0x00000004);
-        }
-        result.frames_ = frames_;
-      } else {
-        result.frames_ = framesBuilder_.build();
-      }
       if (segmentsBuilder_ == null) {
-        if (((bitField0_ & 0x00000008) == 0x00000008)) {
+        if (((bitField0_ & 0x00000002) == 0x00000002)) {
           segments_ = java.util.Collections.unmodifiableList(segments_);
-          bitField0_ = (bitField0_ & ~0x00000008);
+          bitField0_ = (bitField0_ & ~0x00000002);
         }
         result.segments_ = segments_;
       } else {
@@ -644,40 +510,11 @@ public Builder mergeFrom(com.google.cloud.videointelligence.v1p2beta1.TextAnnota
         text_ = other.text_;
         onChanged();
       }
-      if (other.getConfidence() != 0F) {
-        setConfidence(other.getConfidence());
-      }
-      if (framesBuilder_ == null) {
-        if (!other.frames_.isEmpty()) {
-          if (frames_.isEmpty()) {
-            frames_ = other.frames_;
-            bitField0_ = (bitField0_ & ~0x00000004);
-          } else {
-            ensureFramesIsMutable();
-            frames_.addAll(other.frames_);
-          }
-          onChanged();
-        }
-      } else {
-        if (!other.frames_.isEmpty()) {
-          if (framesBuilder_.isEmpty()) {
-            framesBuilder_.dispose();
-            framesBuilder_ = null;
-            frames_ = other.frames_;
-            bitField0_ = (bitField0_ & ~0x00000004);
-            framesBuilder_ = 
-              com.google.protobuf.GeneratedMessageV3.alwaysUseFieldBuilders ?
-                 getFramesFieldBuilder() : null;
-          } else {
-            framesBuilder_.addAllMessages(other.frames_);
-          }
-        }
-      }
       if (segmentsBuilder_ == null) {
         if (!other.segments_.isEmpty()) {
           if (segments_.isEmpty()) {
             segments_ = other.segments_;
-            bitField0_ = (bitField0_ & ~0x00000008);
+            bitField0_ = (bitField0_ & ~0x00000002);
           } else {
             ensureSegmentsIsMutable();
             segments_.addAll(other.segments_);
@@ -690,7 +527,7 @@ public Builder mergeFrom(com.google.cloud.videointelligence.v1p2beta1.TextAnnota
             segmentsBuilder_.dispose();
             segmentsBuilder_ = null;
             segments_ = other.segments_;
-            bitField0_ = (bitField0_ & ~0x00000008);
+            bitField0_ = (bitField0_ & ~0x00000002);
             segmentsBuilder_ = 
               com.google.protobuf.GeneratedMessageV3.alwaysUseFieldBuilders ?
                  getSegmentsFieldBuilder() : null;
@@ -818,365 +655,12 @@ public Builder setTextBytes(
       return this;
     }
 
-    private float confidence_ ;
-    /**
-     * <pre>
-     * Confidence for the track of detected text. It is calculated as the highest
-     * over all frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>float confidence = 2;</code>
-     */
-    public float getConfidence() {
-      return confidence_;
-    }
-    /**
-     * <pre>
-     * Confidence for the track of detected text. It is calculated as the highest
-     * over all frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>float confidence = 2;</code>
-     */
-    public Builder setConfidence(float value) {
-      
-      confidence_ = value;
-      onChanged();
-      return this;
-    }
-    /**
-     * <pre>
-     * Confidence for the track of detected text. It is calculated as the highest
-     * over all frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>float confidence = 2;</code>
-     */
-    public Builder clearConfidence() {
-      
-      confidence_ = 0F;
-      onChanged();
-      return this;
-    }
-
-    private java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> frames_ =
-      java.util.Collections.emptyList();
-    private void ensureFramesIsMutable() {
-      if (!((bitField0_ & 0x00000004) == 0x00000004)) {
-        frames_ = new java.util.ArrayList<com.google.cloud.videointelligence.v1p2beta1.TextFrame>(frames_);
-        bitField0_ |= 0x00000004;
-       }
-    }
-
-    private com.google.protobuf.RepeatedFieldBuilderV3<
-        com.google.cloud.videointelligence.v1p2beta1.TextFrame, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder, com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> framesBuilder_;
-
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> getFramesList() {
-      if (framesBuilder_ == null) {
-        return java.util.Collections.unmodifiableList(frames_);
-      } else {
-        return framesBuilder_.getMessageList();
-      }
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public int getFramesCount() {
-      if (framesBuilder_ == null) {
-        return frames_.size();
-      } else {
-        return framesBuilder_.getCount();
-      }
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public com.google.cloud.videointelligence.v1p2beta1.TextFrame getFrames(int index) {
-      if (framesBuilder_ == null) {
-        return frames_.get(index);
-      } else {
-        return framesBuilder_.getMessage(index);
-      }
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public Builder setFrames(
-        int index, com.google.cloud.videointelligence.v1p2beta1.TextFrame value) {
-      if (framesBuilder_ == null) {
-        if (value == null) {
-          throw new NullPointerException();
-        }
-        ensureFramesIsMutable();
-        frames_.set(index, value);
-        onChanged();
-      } else {
-        framesBuilder_.setMessage(index, value);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public Builder setFrames(
-        int index, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder builderForValue) {
-      if (framesBuilder_ == null) {
-        ensureFramesIsMutable();
-        frames_.set(index, builderForValue.build());
-        onChanged();
-      } else {
-        framesBuilder_.setMessage(index, builderForValue.build());
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public Builder addFrames(com.google.cloud.videointelligence.v1p2beta1.TextFrame value) {
-      if (framesBuilder_ == null) {
-        if (value == null) {
-          throw new NullPointerException();
-        }
-        ensureFramesIsMutable();
-        frames_.add(value);
-        onChanged();
-      } else {
-        framesBuilder_.addMessage(value);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public Builder addFrames(
-        int index, com.google.cloud.videointelligence.v1p2beta1.TextFrame value) {
-      if (framesBuilder_ == null) {
-        if (value == null) {
-          throw new NullPointerException();
-        }
-        ensureFramesIsMutable();
-        frames_.add(index, value);
-        onChanged();
-      } else {
-        framesBuilder_.addMessage(index, value);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public Builder addFrames(
-        com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder builderForValue) {
-      if (framesBuilder_ == null) {
-        ensureFramesIsMutable();
-        frames_.add(builderForValue.build());
-        onChanged();
-      } else {
-        framesBuilder_.addMessage(builderForValue.build());
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public Builder addFrames(
-        int index, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder builderForValue) {
-      if (framesBuilder_ == null) {
-        ensureFramesIsMutable();
-        frames_.add(index, builderForValue.build());
-        onChanged();
-      } else {
-        framesBuilder_.addMessage(index, builderForValue.build());
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public Builder addAllFrames(
-        java.lang.Iterable<? extends com.google.cloud.videointelligence.v1p2beta1.TextFrame> values) {
-      if (framesBuilder_ == null) {
-        ensureFramesIsMutable();
-        com.google.protobuf.AbstractMessageLite.Builder.addAll(
-            values, frames_);
-        onChanged();
-      } else {
-        framesBuilder_.addAllMessages(values);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public Builder clearFrames() {
-      if (framesBuilder_ == null) {
-        frames_ = java.util.Collections.emptyList();
-        bitField0_ = (bitField0_ & ~0x00000004);
-        onChanged();
-      } else {
-        framesBuilder_.clear();
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public Builder removeFrames(int index) {
-      if (framesBuilder_ == null) {
-        ensureFramesIsMutable();
-        frames_.remove(index);
-        onChanged();
-      } else {
-        framesBuilder_.remove(index);
-      }
-      return this;
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder getFramesBuilder(
-        int index) {
-      return getFramesFieldBuilder().getBuilder(index);
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuilder(
-        int index) {
-      if (framesBuilder_ == null) {
-        return frames_.get(index);  } else {
-        return framesBuilder_.getMessageOrBuilder(index);
-      }
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public java.util.List<? extends com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> 
-         getFramesOrBuilderList() {
-      if (framesBuilder_ != null) {
-        return framesBuilder_.getMessageOrBuilderList();
-      } else {
-        return java.util.Collections.unmodifiableList(frames_);
-      }
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder addFramesBuilder() {
-      return getFramesFieldBuilder().addBuilder(
-          com.google.cloud.videointelligence.v1p2beta1.TextFrame.getDefaultInstance());
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder addFramesBuilder(
-        int index) {
-      return getFramesFieldBuilder().addBuilder(
-          index, com.google.cloud.videointelligence.v1p2beta1.TextFrame.getDefaultInstance());
-    }
-    /**
-     * <pre>
-     * Information related to the frames where OCR detected text appears.
-     * </pre>
-     *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-     */
-    public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder> 
-         getFramesBuilderList() {
-      return getFramesFieldBuilder().getBuilderList();
-    }
-    private com.google.protobuf.RepeatedFieldBuilderV3<
-        com.google.cloud.videointelligence.v1p2beta1.TextFrame, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder, com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> 
-        getFramesFieldBuilder() {
-      if (framesBuilder_ == null) {
-        framesBuilder_ = new com.google.protobuf.RepeatedFieldBuilderV3<
-            com.google.cloud.videointelligence.v1p2beta1.TextFrame, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder, com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder>(
-                frames_,
-                ((bitField0_ & 0x00000004) == 0x00000004),
-                getParentForChildren(),
-                isClean());
-        frames_ = null;
-      }
-      return framesBuilder_;
-    }
-
     private java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextSegment> segments_ =
       java.util.Collections.emptyList();
     private void ensureSegmentsIsMutable() {
-      if (!((bitField0_ & 0x00000008) == 0x00000008)) {
+      if (!((bitField0_ & 0x00000002) == 0x00000002)) {
         segments_ = new java.util.ArrayList<com.google.cloud.videointelligence.v1p2beta1.TextSegment>(segments_);
-        bitField0_ |= 0x00000008;
+        bitField0_ |= 0x00000002;
        }
     }
 
@@ -1188,7 +672,7 @@ private void ensureSegmentsIsMutable() {
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextSegment> getSegmentsList() {
       if (segmentsBuilder_ == null) {
@@ -1202,7 +686,7 @@ public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextSegment>
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public int getSegmentsCount() {
       if (segmentsBuilder_ == null) {
@@ -1216,7 +700,7 @@ public int getSegmentsCount() {
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public com.google.cloud.videointelligence.v1p2beta1.TextSegment getSegments(int index) {
       if (segmentsBuilder_ == null) {
@@ -1230,7 +714,7 @@ public com.google.cloud.videointelligence.v1p2beta1.TextSegment getSegments(int
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public Builder setSegments(
         int index, com.google.cloud.videointelligence.v1p2beta1.TextSegment value) {
@@ -1251,7 +735,7 @@ public Builder setSegments(
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public Builder setSegments(
         int index, com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder builderForValue) {
@@ -1269,7 +753,7 @@ public Builder setSegments(
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public Builder addSegments(com.google.cloud.videointelligence.v1p2beta1.TextSegment value) {
       if (segmentsBuilder_ == null) {
@@ -1289,7 +773,7 @@ public Builder addSegments(com.google.cloud.videointelligence.v1p2beta1.TextSegm
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public Builder addSegments(
         int index, com.google.cloud.videointelligence.v1p2beta1.TextSegment value) {
@@ -1310,7 +794,7 @@ public Builder addSegments(
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public Builder addSegments(
         com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder builderForValue) {
@@ -1328,7 +812,7 @@ public Builder addSegments(
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public Builder addSegments(
         int index, com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder builderForValue) {
@@ -1346,7 +830,7 @@ public Builder addSegments(
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public Builder addAllSegments(
         java.lang.Iterable<? extends com.google.cloud.videointelligence.v1p2beta1.TextSegment> values) {
@@ -1365,12 +849,12 @@ public Builder addAllSegments(
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public Builder clearSegments() {
       if (segmentsBuilder_ == null) {
         segments_ = java.util.Collections.emptyList();
-        bitField0_ = (bitField0_ & ~0x00000008);
+        bitField0_ = (bitField0_ & ~0x00000002);
         onChanged();
       } else {
         segmentsBuilder_.clear();
@@ -1382,7 +866,7 @@ public Builder clearSegments() {
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public Builder removeSegments(int index) {
       if (segmentsBuilder_ == null) {
@@ -1399,7 +883,7 @@ public Builder removeSegments(int index) {
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder getSegmentsBuilder(
         int index) {
@@ -1410,7 +894,7 @@ public com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder getSegme
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public com.google.cloud.videointelligence.v1p2beta1.TextSegmentOrBuilder getSegmentsOrBuilder(
         int index) {
@@ -1424,7 +908,7 @@ public com.google.cloud.videointelligence.v1p2beta1.TextSegmentOrBuilder getSegm
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public java.util.List<? extends com.google.cloud.videointelligence.v1p2beta1.TextSegmentOrBuilder> 
          getSegmentsOrBuilderList() {
@@ -1439,7 +923,7 @@ public com.google.cloud.videointelligence.v1p2beta1.TextSegmentOrBuilder getSegm
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder addSegmentsBuilder() {
       return getSegmentsFieldBuilder().addBuilder(
@@ -1450,7 +934,7 @@ public com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder addSegme
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder addSegmentsBuilder(
         int index) {
@@ -1462,7 +946,7 @@ public com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder addSegme
      * All video segments where OCR detected text appears.
      * </pre>
      *
-     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
      */
     public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder> 
          getSegmentsBuilderList() {
@@ -1475,7 +959,7 @@ public com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder addSegme
         segmentsBuilder_ = new com.google.protobuf.RepeatedFieldBuilderV3<
             com.google.cloud.videointelligence.v1p2beta1.TextSegment, com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder, com.google.cloud.videointelligence.v1p2beta1.TextSegmentOrBuilder>(
                 segments_,
-                ((bitField0_ & 0x00000008) == 0x00000008),
+                ((bitField0_ & 0x00000002) == 0x00000002),
                 getParentForChildren(),
                 isClean());
         segments_ = null;
diff --git a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextAnnotationOrBuilder.java b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextAnnotationOrBuilder.java
index c9f1f91600eb..55860a35eeb3 100644
--- a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextAnnotationOrBuilder.java
+++ b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextAnnotationOrBuilder.java
@@ -25,66 +25,12 @@ public interface TextAnnotationOrBuilder extends
   com.google.protobuf.ByteString
       getTextBytes();
 
-  /**
-   * <pre>
-   * Confidence for the track of detected text. It is calculated as the highest
-   * over all frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>float confidence = 2;</code>
-   */
-  float getConfidence();
-
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> 
-      getFramesList();
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  com.google.cloud.videointelligence.v1p2beta1.TextFrame getFrames(int index);
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  int getFramesCount();
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  java.util.List<? extends com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> 
-      getFramesOrBuilderList();
-  /**
-   * <pre>
-   * Information related to the frames where OCR detected text appears.
-   * </pre>
-   *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
-   */
-  com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuilder(
-      int index);
-
   /**
    * <pre>
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextSegment> 
       getSegmentsList();
@@ -93,7 +39,7 @@ com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuild
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   com.google.cloud.videointelligence.v1p2beta1.TextSegment getSegments(int index);
   /**
@@ -101,7 +47,7 @@ com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuild
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   int getSegmentsCount();
   /**
@@ -109,7 +55,7 @@ com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuild
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   java.util.List<? extends com.google.cloud.videointelligence.v1p2beta1.TextSegmentOrBuilder> 
       getSegmentsOrBuilderList();
@@ -118,7 +64,7 @@ com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuild
    * All video segments where OCR detected text appears.
    * </pre>
    *
-   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 4;</code>
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextSegment segments = 2;</code>
    */
   com.google.cloud.videointelligence.v1p2beta1.TextSegmentOrBuilder getSegmentsOrBuilder(
       int index);
diff --git a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextSegment.java b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextSegment.java
index 381264338009..7c71650e1a4e 100644
--- a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextSegment.java
+++ b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextSegment.java
@@ -20,6 +20,8 @@ private TextSegment(com.google.protobuf.GeneratedMessageV3.Builder<?> builder) {
     super(builder);
   }
   private TextSegment() {
+    confidence_ = 0F;
+    frames_ = java.util.Collections.emptyList();
   }
 
   @java.lang.Override
@@ -59,6 +61,20 @@ private TextSegment(
 
             break;
           }
+          case 21: {
+
+            confidence_ = input.readFloat();
+            break;
+          }
+          case 26: {
+            if (!((mutable_bitField0_ & 0x00000004) == 0x00000004)) {
+              frames_ = new java.util.ArrayList<com.google.cloud.videointelligence.v1p2beta1.TextFrame>();
+              mutable_bitField0_ |= 0x00000004;
+            }
+            frames_.add(
+                input.readMessage(com.google.cloud.videointelligence.v1p2beta1.TextFrame.parser(), extensionRegistry));
+            break;
+          }
           default: {
             if (!parseUnknownFieldProto3(
                 input, unknownFields, extensionRegistry, tag)) {
@@ -74,6 +90,9 @@ private TextSegment(
       throw new com.google.protobuf.InvalidProtocolBufferException(
           e).setUnfinishedMessage(this);
     } finally {
+      if (((mutable_bitField0_ & 0x00000004) == 0x00000004)) {
+        frames_ = java.util.Collections.unmodifiableList(frames_);
+      }
       this.unknownFields = unknownFields.build();
       makeExtensionsImmutable();
     }
@@ -91,6 +110,7 @@ private TextSegment(
             com.google.cloud.videointelligence.v1p2beta1.TextSegment.class, com.google.cloud.videointelligence.v1p2beta1.TextSegment.Builder.class);
   }
 
+  private int bitField0_;
   public static final int SEGMENT_FIELD_NUMBER = 1;
   private com.google.cloud.videointelligence.v1p2beta1.VideoSegment segment_;
   /**
@@ -124,6 +144,75 @@ public com.google.cloud.videointelligence.v1p2beta1.VideoSegmentOrBuilder getSeg
     return getSegment();
   }
 
+  public static final int CONFIDENCE_FIELD_NUMBER = 2;
+  private float confidence_;
+  /**
+   * <pre>
+   * Confidence for the track of detected text. It is calculated as the highest
+   * over all frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>float confidence = 2;</code>
+   */
+  public float getConfidence() {
+    return confidence_;
+  }
+
+  public static final int FRAMES_FIELD_NUMBER = 3;
+  private java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> frames_;
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> getFramesList() {
+    return frames_;
+  }
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  public java.util.List<? extends com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> 
+      getFramesOrBuilderList() {
+    return frames_;
+  }
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  public int getFramesCount() {
+    return frames_.size();
+  }
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  public com.google.cloud.videointelligence.v1p2beta1.TextFrame getFrames(int index) {
+    return frames_.get(index);
+  }
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  public com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuilder(
+      int index) {
+    return frames_.get(index);
+  }
+
   private byte memoizedIsInitialized = -1;
   @java.lang.Override
   public final boolean isInitialized() {
@@ -141,6 +230,12 @@ public void writeTo(com.google.protobuf.CodedOutputStream output)
     if (segment_ != null) {
       output.writeMessage(1, getSegment());
     }
+    if (confidence_ != 0F) {
+      output.writeFloat(2, confidence_);
+    }
+    for (int i = 0; i < frames_.size(); i++) {
+      output.writeMessage(3, frames_.get(i));
+    }
     unknownFields.writeTo(output);
   }
 
@@ -154,6 +249,14 @@ public int getSerializedSize() {
       size += com.google.protobuf.CodedOutputStream
         .computeMessageSize(1, getSegment());
     }
+    if (confidence_ != 0F) {
+      size += com.google.protobuf.CodedOutputStream
+        .computeFloatSize(2, confidence_);
+    }
+    for (int i = 0; i < frames_.size(); i++) {
+      size += com.google.protobuf.CodedOutputStream
+        .computeMessageSize(3, frames_.get(i));
+    }
     size += unknownFields.getSerializedSize();
     memoizedSize = size;
     return size;
@@ -175,6 +278,12 @@ public boolean equals(final java.lang.Object obj) {
       result = result && getSegment()
           .equals(other.getSegment());
     }
+    result = result && (
+        java.lang.Float.floatToIntBits(getConfidence())
+        == java.lang.Float.floatToIntBits(
+            other.getConfidence()));
+    result = result && getFramesList()
+        .equals(other.getFramesList());
     result = result && unknownFields.equals(other.unknownFields);
     return result;
   }
@@ -190,6 +299,13 @@ public int hashCode() {
       hash = (37 * hash) + SEGMENT_FIELD_NUMBER;
       hash = (53 * hash) + getSegment().hashCode();
     }
+    hash = (37 * hash) + CONFIDENCE_FIELD_NUMBER;
+    hash = (53 * hash) + java.lang.Float.floatToIntBits(
+        getConfidence());
+    if (getFramesCount() > 0) {
+      hash = (37 * hash) + FRAMES_FIELD_NUMBER;
+      hash = (53 * hash) + getFramesList().hashCode();
+    }
     hash = (29 * hash) + unknownFields.hashCode();
     memoizedHashCode = hash;
     return hash;
@@ -322,6 +438,7 @@ private Builder(
     private void maybeForceBuilderInitialization() {
       if (com.google.protobuf.GeneratedMessageV3
               .alwaysUseFieldBuilders) {
+        getFramesFieldBuilder();
       }
     }
     @java.lang.Override
@@ -333,6 +450,14 @@ public Builder clear() {
         segment_ = null;
         segmentBuilder_ = null;
       }
+      confidence_ = 0F;
+
+      if (framesBuilder_ == null) {
+        frames_ = java.util.Collections.emptyList();
+        bitField0_ = (bitField0_ & ~0x00000004);
+      } else {
+        framesBuilder_.clear();
+      }
       return this;
     }
 
@@ -359,11 +484,24 @@ public com.google.cloud.videointelligence.v1p2beta1.TextSegment build() {
     @java.lang.Override
     public com.google.cloud.videointelligence.v1p2beta1.TextSegment buildPartial() {
       com.google.cloud.videointelligence.v1p2beta1.TextSegment result = new com.google.cloud.videointelligence.v1p2beta1.TextSegment(this);
+      int from_bitField0_ = bitField0_;
+      int to_bitField0_ = 0;
       if (segmentBuilder_ == null) {
         result.segment_ = segment_;
       } else {
         result.segment_ = segmentBuilder_.build();
       }
+      result.confidence_ = confidence_;
+      if (framesBuilder_ == null) {
+        if (((bitField0_ & 0x00000004) == 0x00000004)) {
+          frames_ = java.util.Collections.unmodifiableList(frames_);
+          bitField0_ = (bitField0_ & ~0x00000004);
+        }
+        result.frames_ = frames_;
+      } else {
+        result.frames_ = framesBuilder_.build();
+      }
+      result.bitField0_ = to_bitField0_;
       onBuilt();
       return result;
     }
@@ -415,6 +553,35 @@ public Builder mergeFrom(com.google.cloud.videointelligence.v1p2beta1.TextSegmen
       if (other.hasSegment()) {
         mergeSegment(other.getSegment());
       }
+      if (other.getConfidence() != 0F) {
+        setConfidence(other.getConfidence());
+      }
+      if (framesBuilder_ == null) {
+        if (!other.frames_.isEmpty()) {
+          if (frames_.isEmpty()) {
+            frames_ = other.frames_;
+            bitField0_ = (bitField0_ & ~0x00000004);
+          } else {
+            ensureFramesIsMutable();
+            frames_.addAll(other.frames_);
+          }
+          onChanged();
+        }
+      } else {
+        if (!other.frames_.isEmpty()) {
+          if (framesBuilder_.isEmpty()) {
+            framesBuilder_.dispose();
+            framesBuilder_ = null;
+            frames_ = other.frames_;
+            bitField0_ = (bitField0_ & ~0x00000004);
+            framesBuilder_ = 
+              com.google.protobuf.GeneratedMessageV3.alwaysUseFieldBuilders ?
+                 getFramesFieldBuilder() : null;
+          } else {
+            framesBuilder_.addAllMessages(other.frames_);
+          }
+        }
+      }
       this.mergeUnknownFields(other.unknownFields);
       onChanged();
       return this;
@@ -443,6 +610,7 @@ public Builder mergeFrom(
       }
       return this;
     }
+    private int bitField0_;
 
     private com.google.cloud.videointelligence.v1p2beta1.VideoSegment segment_ = null;
     private com.google.protobuf.SingleFieldBuilderV3<
@@ -596,6 +764,359 @@ public com.google.cloud.videointelligence.v1p2beta1.VideoSegmentOrBuilder getSeg
       }
       return segmentBuilder_;
     }
+
+    private float confidence_ ;
+    /**
+     * <pre>
+     * Confidence for the track of detected text. It is calculated as the highest
+     * over all frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>float confidence = 2;</code>
+     */
+    public float getConfidence() {
+      return confidence_;
+    }
+    /**
+     * <pre>
+     * Confidence for the track of detected text. It is calculated as the highest
+     * over all frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>float confidence = 2;</code>
+     */
+    public Builder setConfidence(float value) {
+      
+      confidence_ = value;
+      onChanged();
+      return this;
+    }
+    /**
+     * <pre>
+     * Confidence for the track of detected text. It is calculated as the highest
+     * over all frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>float confidence = 2;</code>
+     */
+    public Builder clearConfidence() {
+      
+      confidence_ = 0F;
+      onChanged();
+      return this;
+    }
+
+    private java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> frames_ =
+      java.util.Collections.emptyList();
+    private void ensureFramesIsMutable() {
+      if (!((bitField0_ & 0x00000004) == 0x00000004)) {
+        frames_ = new java.util.ArrayList<com.google.cloud.videointelligence.v1p2beta1.TextFrame>(frames_);
+        bitField0_ |= 0x00000004;
+       }
+    }
+
+    private com.google.protobuf.RepeatedFieldBuilderV3<
+        com.google.cloud.videointelligence.v1p2beta1.TextFrame, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder, com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> framesBuilder_;
+
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> getFramesList() {
+      if (framesBuilder_ == null) {
+        return java.util.Collections.unmodifiableList(frames_);
+      } else {
+        return framesBuilder_.getMessageList();
+      }
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public int getFramesCount() {
+      if (framesBuilder_ == null) {
+        return frames_.size();
+      } else {
+        return framesBuilder_.getCount();
+      }
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public com.google.cloud.videointelligence.v1p2beta1.TextFrame getFrames(int index) {
+      if (framesBuilder_ == null) {
+        return frames_.get(index);
+      } else {
+        return framesBuilder_.getMessage(index);
+      }
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public Builder setFrames(
+        int index, com.google.cloud.videointelligence.v1p2beta1.TextFrame value) {
+      if (framesBuilder_ == null) {
+        if (value == null) {
+          throw new NullPointerException();
+        }
+        ensureFramesIsMutable();
+        frames_.set(index, value);
+        onChanged();
+      } else {
+        framesBuilder_.setMessage(index, value);
+      }
+      return this;
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public Builder setFrames(
+        int index, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder builderForValue) {
+      if (framesBuilder_ == null) {
+        ensureFramesIsMutable();
+        frames_.set(index, builderForValue.build());
+        onChanged();
+      } else {
+        framesBuilder_.setMessage(index, builderForValue.build());
+      }
+      return this;
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public Builder addFrames(com.google.cloud.videointelligence.v1p2beta1.TextFrame value) {
+      if (framesBuilder_ == null) {
+        if (value == null) {
+          throw new NullPointerException();
+        }
+        ensureFramesIsMutable();
+        frames_.add(value);
+        onChanged();
+      } else {
+        framesBuilder_.addMessage(value);
+      }
+      return this;
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public Builder addFrames(
+        int index, com.google.cloud.videointelligence.v1p2beta1.TextFrame value) {
+      if (framesBuilder_ == null) {
+        if (value == null) {
+          throw new NullPointerException();
+        }
+        ensureFramesIsMutable();
+        frames_.add(index, value);
+        onChanged();
+      } else {
+        framesBuilder_.addMessage(index, value);
+      }
+      return this;
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public Builder addFrames(
+        com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder builderForValue) {
+      if (framesBuilder_ == null) {
+        ensureFramesIsMutable();
+        frames_.add(builderForValue.build());
+        onChanged();
+      } else {
+        framesBuilder_.addMessage(builderForValue.build());
+      }
+      return this;
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public Builder addFrames(
+        int index, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder builderForValue) {
+      if (framesBuilder_ == null) {
+        ensureFramesIsMutable();
+        frames_.add(index, builderForValue.build());
+        onChanged();
+      } else {
+        framesBuilder_.addMessage(index, builderForValue.build());
+      }
+      return this;
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public Builder addAllFrames(
+        java.lang.Iterable<? extends com.google.cloud.videointelligence.v1p2beta1.TextFrame> values) {
+      if (framesBuilder_ == null) {
+        ensureFramesIsMutable();
+        com.google.protobuf.AbstractMessageLite.Builder.addAll(
+            values, frames_);
+        onChanged();
+      } else {
+        framesBuilder_.addAllMessages(values);
+      }
+      return this;
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public Builder clearFrames() {
+      if (framesBuilder_ == null) {
+        frames_ = java.util.Collections.emptyList();
+        bitField0_ = (bitField0_ & ~0x00000004);
+        onChanged();
+      } else {
+        framesBuilder_.clear();
+      }
+      return this;
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public Builder removeFrames(int index) {
+      if (framesBuilder_ == null) {
+        ensureFramesIsMutable();
+        frames_.remove(index);
+        onChanged();
+      } else {
+        framesBuilder_.remove(index);
+      }
+      return this;
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder getFramesBuilder(
+        int index) {
+      return getFramesFieldBuilder().getBuilder(index);
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuilder(
+        int index) {
+      if (framesBuilder_ == null) {
+        return frames_.get(index);  } else {
+        return framesBuilder_.getMessageOrBuilder(index);
+      }
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public java.util.List<? extends com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> 
+         getFramesOrBuilderList() {
+      if (framesBuilder_ != null) {
+        return framesBuilder_.getMessageOrBuilderList();
+      } else {
+        return java.util.Collections.unmodifiableList(frames_);
+      }
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder addFramesBuilder() {
+      return getFramesFieldBuilder().addBuilder(
+          com.google.cloud.videointelligence.v1p2beta1.TextFrame.getDefaultInstance());
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder addFramesBuilder(
+        int index) {
+      return getFramesFieldBuilder().addBuilder(
+          index, com.google.cloud.videointelligence.v1p2beta1.TextFrame.getDefaultInstance());
+    }
+    /**
+     * <pre>
+     * Information related to the frames where OCR detected text appears.
+     * </pre>
+     *
+     * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+     */
+    public java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder> 
+         getFramesBuilderList() {
+      return getFramesFieldBuilder().getBuilderList();
+    }
+    private com.google.protobuf.RepeatedFieldBuilderV3<
+        com.google.cloud.videointelligence.v1p2beta1.TextFrame, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder, com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> 
+        getFramesFieldBuilder() {
+      if (framesBuilder_ == null) {
+        framesBuilder_ = new com.google.protobuf.RepeatedFieldBuilderV3<
+            com.google.cloud.videointelligence.v1p2beta1.TextFrame, com.google.cloud.videointelligence.v1p2beta1.TextFrame.Builder, com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder>(
+                frames_,
+                ((bitField0_ & 0x00000004) == 0x00000004),
+                getParentForChildren(),
+                isClean());
+        frames_ = null;
+      }
+      return framesBuilder_;
+    }
     @java.lang.Override
     public final Builder setUnknownFields(
         final com.google.protobuf.UnknownFieldSet unknownFields) {
diff --git a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextSegmentOrBuilder.java b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextSegmentOrBuilder.java
index 6334f77c3737..a39e239caa12 100644
--- a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextSegmentOrBuilder.java
+++ b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/TextSegmentOrBuilder.java
@@ -31,4 +31,58 @@ public interface TextSegmentOrBuilder extends
    * <code>.google.cloud.videointelligence.v1p2beta1.VideoSegment segment = 1;</code>
    */
   com.google.cloud.videointelligence.v1p2beta1.VideoSegmentOrBuilder getSegmentOrBuilder();
+
+  /**
+   * <pre>
+   * Confidence for the track of detected text. It is calculated as the highest
+   * over all frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>float confidence = 2;</code>
+   */
+  float getConfidence();
+
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  java.util.List<com.google.cloud.videointelligence.v1p2beta1.TextFrame> 
+      getFramesList();
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  com.google.cloud.videointelligence.v1p2beta1.TextFrame getFrames(int index);
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  int getFramesCount();
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  java.util.List<? extends com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder> 
+      getFramesOrBuilderList();
+  /**
+   * <pre>
+   * Information related to the frames where OCR detected text appears.
+   * </pre>
+   *
+   * <code>repeated .google.cloud.videointelligence.v1p2beta1.TextFrame frames = 3;</code>
+   */
+  com.google.cloud.videointelligence.v1p2beta1.TextFrameOrBuilder getFramesOrBuilder(
+      int index);
 }
diff --git a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/VideoIntelligenceServiceProto.java b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/VideoIntelligenceServiceProto.java
index 3c70cb97ae47..0744f6ed6445 100644
--- a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/VideoIntelligenceServiceProto.java
+++ b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/java/com/google/cloud/videointelligence/v1p2beta1/VideoIntelligenceServiceProto.java
@@ -239,16 +239,16 @@ public static void registerAllExtensions(
       "\n\001x\030\001 \001(\002\022\t\n\001y\030\002 \001(\002\"f\n\026NormalizedBoundi" +
       "ngPoly\022L\n\010vertices\030\001 \003(\0132:.google.cloud." +
       "videointelligence.v1p2beta1.NormalizedVe" +
-      "rtex\"V\n\013TextSegment\022G\n\007segment\030\001 \001(\01326.g" +
-      "oogle.cloud.videointelligence.v1p2beta1." +
-      "VideoSegment\"\233\001\n\tTextFrame\022^\n\024rotated_bo" +
-      "unding_box\030\001 \001(\0132@.google.cloud.videoint" +
-      "elligence.v1p2beta1.NormalizedBoundingPo" +
-      "ly\022.\n\013time_offset\030\002 \001(\0132\031.google.protobu" +
-      "f.Duration\"\300\001\n\016TextAnnotation\022\014\n\004text\030\001 " +
-      "\001(\t\022\022\n\nconfidence\030\002 \001(\002\022C\n\006frames\030\003 \003(\0132" +
-      "3.google.cloud.videointelligence.v1p2bet" +
-      "a1.TextFrame\022G\n\010segments\030\004 \003(\01325.google." +
+      "rtex\"\257\001\n\013TextSegment\022G\n\007segment\030\001 \001(\01326." +
+      "google.cloud.videointelligence.v1p2beta1" +
+      ".VideoSegment\022\022\n\nconfidence\030\002 \001(\002\022C\n\006fra" +
+      "mes\030\003 \003(\01323.google.cloud.videointelligen" +
+      "ce.v1p2beta1.TextFrame\"\233\001\n\tTextFrame\022^\n\024" +
+      "rotated_bounding_box\030\001 \001(\0132@.google.clou" +
+      "d.videointelligence.v1p2beta1.Normalized" +
+      "BoundingPoly\022.\n\013time_offset\030\002 \001(\0132\031.goog" +
+      "le.protobuf.Duration\"g\n\016TextAnnotation\022\014" +
+      "\n\004text\030\001 \001(\t\022G\n\010segments\030\002 \003(\01325.google." +
       "cloud.videointelligence.v1p2beta1.TextSe" +
       "gment\"\247\001\n\023ObjectTrackingFrame\022`\n\027normali" +
       "zed_bounding_box\030\001 \001(\0132?.google.cloud.vi" +
@@ -426,7 +426,7 @@ public com.google.protobuf.ExtensionRegistry assignDescriptors(
     internal_static_google_cloud_videointelligence_v1p2beta1_TextSegment_fieldAccessorTable = new
       com.google.protobuf.GeneratedMessageV3.FieldAccessorTable(
         internal_static_google_cloud_videointelligence_v1p2beta1_TextSegment_descriptor,
-        new java.lang.String[] { "Segment", });
+        new java.lang.String[] { "Segment", "Confidence", "Frames", });
     internal_static_google_cloud_videointelligence_v1p2beta1_TextFrame_descriptor =
       getDescriptor().getMessageTypes().get(21);
     internal_static_google_cloud_videointelligence_v1p2beta1_TextFrame_fieldAccessorTable = new
@@ -438,7 +438,7 @@ public com.google.protobuf.ExtensionRegistry assignDescriptors(
     internal_static_google_cloud_videointelligence_v1p2beta1_TextAnnotation_fieldAccessorTable = new
       com.google.protobuf.GeneratedMessageV3.FieldAccessorTable(
         internal_static_google_cloud_videointelligence_v1p2beta1_TextAnnotation_descriptor,
-        new java.lang.String[] { "Text", "Confidence", "Frames", "Segments", });
+        new java.lang.String[] { "Text", "Segments", });
     internal_static_google_cloud_videointelligence_v1p2beta1_ObjectTrackingFrame_descriptor =
       getDescriptor().getMessageTypes().get(23);
     internal_static_google_cloud_videointelligence_v1p2beta1_ObjectTrackingFrame_fieldAccessorTable = new
diff --git a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/proto/google/cloud/videointelligence/v1p2beta1/video_intelligence.proto b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/proto/google/cloud/videointelligence/v1p2beta1/video_intelligence.proto
index 2c1af2efddd7..ce540c90c082 100644
--- a/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/proto/google/cloud/videointelligence/v1p2beta1/video_intelligence.proto
+++ b/google-api-grpc/proto-google-cloud-video-intelligence-v1p2beta1/src/main/proto/google/cloud/videointelligence/v1p2beta1/video_intelligence.proto
@@ -352,6 +352,13 @@ message NormalizedBoundingPoly {
 message TextSegment {
   // Video segment where a text snippet was detected.
   VideoSegment segment = 1;
+
+  // Confidence for the track of detected text. It is calculated as the highest
+  // over all frames where OCR detected text appears.
+  float confidence = 2;
+
+  // Information related to the frames where OCR detected text appears.
+  repeated TextFrame frames = 3;
 }
 
 // Video frame level annotation results for text annotation (OCR).
@@ -372,15 +379,8 @@ message TextAnnotation {
   // The detected text.
   string text = 1;
 
-  // Confidence for the track of detected text. It is calculated as the highest
-  // over all frames where OCR detected text appears.
-  float confidence = 2;
-
-  // Information related to the frames where OCR detected text appears.
-  repeated TextFrame frames = 3;
-
   // All video segments where OCR detected text appears.
-  repeated TextSegment segments = 4;
+  repeated TextSegment segments = 2;
 }
 
 // Video frame level annotations for object detection and tracking. This field