Add support for AIS batch loading for ASR audio processing#15102
Merged
tbartley94 merged 1 commit intoNVIDIA-NeMo:mainfrom Dec 2, 2025
Merged
Add support for AIS batch loading for ASR audio processing#15102tbartley94 merged 1 commit intoNVIDIA-NeMo:mainfrom
tbartley94 merged 1 commit intoNVIDIA-NeMo:mainfrom
Conversation
gaikwadabhishek
force-pushed
the
get-batch
branch
from
47ef8e6 to
cbae6c3
Compare
pzelasko
reviewed
Nov 21, 2025
gaikwadabhishek
force-pushed
the
get-batch
branch
3 times, most recently
from
6ae836b to
3e60dd6
Compare
gaikwadabhishek
commented
Nov 21, 2025
Contributor
Author
|
tests are failing due to missing lhotse release. will re-run after the new lhotse release is in place |
auto-merge was automatically disabled
November 24, 2025 17:51
Head branch was pushed to by a user without write access
gaikwadabhishek
force-pushed
the
get-batch
branch
from
3e60dd6 to
18557ae
Compare
* Introduced environment variable USE_AIS_GET_BATCH to toggle AIS batch loading. * Updated PromptedAudioToTextLhotseDataset to utilize AISBatchLoader when enabled. * Modified LazyNeMoTarredIterator to handle URL-based recordings when AIS batch loading is active. Implement URL-based audio loading using AIStore's Get-Batch API to improve data pipeline efficiency. This allows batch fetching of multiple audio files without local tar archive extraction, offloading processing to AIStore. Signed-off-by: Abhishek Gaikwad <gaikwadabhishek1997@gmail.com>
gaikwadabhishek
force-pushed
the
get-batch
branch
from
18557ae to
c0202a2
Compare
shababo
pushed a commit
to modal-projects/NeMo
that referenced
this pull request
Dec 4, 2025
…Mo#15102) * Introduced environment variable USE_AIS_GET_BATCH to toggle AIS batch loading. * Updated PromptedAudioToTextLhotseDataset to utilize AISBatchLoader when enabled. * Modified LazyNeMoTarredIterator to handle URL-based recordings when AIS batch loading is active. Implement URL-based audio loading using AIStore's Get-Batch API to improve data pipeline efficiency. This allows batch fetching of multiple audio files without local tar archive extraction, offloading processing to AIStore. Signed-off-by: Abhishek Gaikwad <gaikwadabhishek1997@gmail.com> Signed-off-by: shababo <ben.shababo@modal.com>
shababo
pushed a commit
to modal-projects/NeMo
that referenced
this pull request
Dec 4, 2025
…Mo#15102) * Introduced environment variable USE_AIS_GET_BATCH to toggle AIS batch loading. * Updated PromptedAudioToTextLhotseDataset to utilize AISBatchLoader when enabled. * Modified LazyNeMoTarredIterator to handle URL-based recordings when AIS batch loading is active. Implement URL-based audio loading using AIStore's Get-Batch API to improve data pipeline efficiency. This allows batch fetching of multiple audio files without local tar archive extraction, offloading processing to AIStore. Signed-off-by: Abhishek Gaikwad <gaikwadabhishek1997@gmail.com> Signed-off-by: shababo <ben.shababo@modal.com>
quapham
pushed a commit
to quapham/NeMo
that referenced
this pull request
Dec 16, 2025
…Mo#15102) * Introduced environment variable USE_AIS_GET_BATCH to toggle AIS batch loading. * Updated PromptedAudioToTextLhotseDataset to utilize AISBatchLoader when enabled. * Modified LazyNeMoTarredIterator to handle URL-based recordings when AIS batch loading is active. Implement URL-based audio loading using AIStore's Get-Batch API to improve data pipeline efficiency. This allows batch fetching of multiple audio files without local tar archive extraction, offloading processing to AIStore. Signed-off-by: Abhishek Gaikwad <gaikwadabhishek1997@gmail.com> Signed-off-by: quanpham <youngkwan199@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do ?
Implement URL-based audio loading using AIStore's Get-Batch API to improve data pipeline efficiency. This allows batch fetching of multiple audio files without local tar archive extraction, offloading processing to AIStore.
Collection: asr
Changelog
Usage
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:
PR Type:
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