GPU-accelerated Phrase-Boosting (GPU-PB) for AED decoding

[andrusenkoau]

Important

The Update branch button must only be pressed in very rare occassions.
An outdated branch is never blocking the merge of a PR.
Please reach out to the automation team before pressing that button.

What does this PR do?

Implementation of GPU-accelerated Phrase-Boosting method (GPU-PB) in shallow fusion mode for AED decoding (e.g. Canary) based on NGPU-LM infrastructure. The method supports greedy (beam=1) and beam search decoding. GPU-PB is compatible with NGPU-LM at the same decoding run.

Collection: [ASR]

Changelog

• Add specific line by line info of high level changes in this PR.

Usage

Step 1 (Optional): Build boosting tree for the specific ASR model
You can skip this step and provide a context file or a list with key phrases directly to speech_to_text_eval.py (step 2)

python scripts/asr_context_biasing/build_gpu_boosting_tree.py \
        asr_model_path=${ASR_NEMO_MODEL_FILE} \
        key_phrases_file=${CONTEXT_BIASING_LIST} \
        save_to=${PATH_TO_SAVE_BTREE} \
        context_score=${CONTEXT_SCORE} \
        depth_scaling=${DEPTH_SCALING} \
        use_triton=True

Step 2: Run AED decoding with boosting tree

python examples/asr/speech_to_text_eval.py \
        model_path=${MODEL_NAME} \
        dataset_manifest=${EVAL_MANIFEST} \
        batch_size=${BATCH_SIZE} \
        output_filename=${OUT_MANIFEST} \
        multitask_decoding.strategy="beam" \
        multitask_decoding.beam.beam_size=${BEAM_SIZE} \
        multitask_decoding.beam.boosting_tree.key_phrases_file=${CONTEXT_BIASING_LIST} \
        multitask_decoding.beam.boosting_tree.context_score=1.0 \
        multitask_decoding.beam.boosting_tree.depth_scaling=1.0 \
        multitask_decoding.beam.boosting_tree_alpha=${BT_ALPHA} \
        gt_lang_attr_name="target_lang" \
        gt_text_attr_name="text" \
        +prompt.pnc=no \
        +prompt.task=asr \
        +prompt.source_lang=en \
        +prompt.target_lang=en \

Step 3: Evaluate F-score for context phrases

python scripts/asr_context_biasing/compute_key_words_fscore.py \
        --input_manifest=${DECODING_MANIFEST} \
        --key_words_file=${CONTEXT_PHRASES_LIST}

GitHub Actions CI

The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.

The GitHub Actions CI will run automatically when the "Run CICD" label is added to the PR.
To re-run CI remove and add the label again.
To run CI on an untrusted fork, a NeMo user with write access must first click "Approve and run".

Before your PR is "Ready for review"

Pre checks:

• Make sure you read and followed Contributor guidelines
• Did you write any new necessary tests?
• Did you add or update any necessary documentation?
• Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
  • Reviewer: Does the PR have correct import guards for all optional libraries?

PR Type:

• New Feature
• Bugfix
• Documentation

If you haven't finished some of the above items you can still open "Draft" PR.

Who can review?

Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.

Additional Information

• Related to # (issue)

[artbataev]

Thanks a lot! Great work!