[multimodal] Let Audio take float data blob #14427
Merged
Conversation
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
If the processed audio went through Mel transform, the spectrogram are float values. We should allow `Audio` class to be able to take this, since multimodal runner pybind API will have to be able to take processed input. Once we have the pybind API we can do something like: ```python model_id = "mistralai/Voxtral-Mini-3B-2507" processor = AutoProcessor.from_pretrained(model_id) audio_url = "https://huggingface.co/datasets/eustlb/audio-samples/resolve/main/dude_where_is_my_car.wav" conversation = [ { "role": "user", "content": [ {"type": "audio", "url": audio_url}, { "type": "text", "text": "What can you tell me about this audio?", }, ], }, ] inputs = processor.apply_chat_template(conversation, tokenize=True, return_dict=True, return_tensors="pt") inputs_combined = [ make_text_input("<s>[INST][BEGIN_AUDIO]"), make_audio_input(inputs["input_features"]), make_text_input("\nWhat can you tell me about this audio?[/INST]"), ] runner = MultimodalRunner("voxtral.pte", "tekken.json", None) config = GenerationConfig() config.max_new_tokens = 100 runner.generate(inputs_combined, config) ```
larryliu0820
requested review from
jackzhxng,
lucylq,
mergennachin and
swolchok
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/14427
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 3 PendingAs of commit 8b6a3d3 with merge base 0f22062 ( NEW FAILURE - The following job has failed:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
jackzhxng
approved these changes
Sep 19, 2025
StrycekSimon
pushed a commit
to nxp-upstream/executorch
that referenced
this pull request
Sep 23, 2025
If the processed audio went through Mel transform, the spectrogram are float values. We should allow `Audio` class to be able to take this, since multimodal runner pybind API will have to be able to take processed input. Once we have the pybind API we can do something like: ```python model_id = "mistralai/Voxtral-Mini-3B-2507" processor = AutoProcessor.from_pretrained(model_id) audio_url = "https://huggingface.co/datasets/eustlb/audio-samples/resolve/main/dude_where_is_my_car.wav" conversation = [ { "role": "user", "content": [ {"type": "audio", "url": audio_url}, { "type": "text", "text": "What can you tell me about this audio?", }, ], }, ] inputs = processor.apply_chat_template(conversation, tokenize=True, return_dict=True, return_tensors="pt") inputs_combined = [ make_text_input("<s>[INST][BEGIN_AUDIO]"), make_audio_input(inputs["input_features"]), make_text_input("\nWhat can you tell me about this audio?[/INST]"), ] runner = MultimodalRunner("voxtral.pte", "tekken.json", None) config = GenerationConfig() config.max_new_tokens = 100 runner.generate(inputs_combined, config) ``` ### Summary [PLEASE REMOVE] See [CONTRIBUTING.md's Pull Requests](https://github.com/pytorch/executorch/blob/main/CONTRIBUTING.md#pull-requests) for ExecuTorch PR guidelines. [PLEASE REMOVE] If this PR closes an issue, please add a `Fixes #<issue-id>` line. [PLEASE REMOVE] If this PR introduces a fix or feature that should be the upcoming release notes, please add a "Release notes: <area>" label. For a list of available release notes labels, check out [CONTRIBUTING.md's Pull Requests](https://github.com/pytorch/executorch/blob/main/CONTRIBUTING.md#pull-requests). ### Test plan [PLEASE REMOVE] How did you test this PR? Please write down any manual commands you used and note down tests that you have written if applicable.
|
@pytorchbot cherry-pick --onto release/1.0 -c critical |
pytorchbot
pushed a commit
that referenced
this pull request
Oct 9, 2025
If the processed audio went through Mel transform, the spectrogram are float values. We should allow `Audio` class to be able to take this, since multimodal runner pybind API will have to be able to take processed input. Once we have the pybind API we can do something like: ```python model_id = "mistralai/Voxtral-Mini-3B-2507" processor = AutoProcessor.from_pretrained(model_id) audio_url = "https://huggingface.co/datasets/eustlb/audio-samples/resolve/main/dude_where_is_my_car.wav" conversation = [ { "role": "user", "content": [ {"type": "audio", "url": audio_url}, { "type": "text", "text": "What can you tell me about this audio?", }, ], }, ] inputs = processor.apply_chat_template(conversation, tokenize=True, return_dict=True, return_tensors="pt") inputs_combined = [ make_text_input("<s>[INST][BEGIN_AUDIO]"), make_audio_input(inputs["input_features"]), make_text_input("\nWhat can you tell me about this audio?[/INST]"), ] runner = MultimodalRunner("voxtral.pte", "tekken.json", None) config = GenerationConfig() config.max_new_tokens = 100 runner.generate(inputs_combined, config) ``` ### Summary [PLEASE REMOVE] See [CONTRIBUTING.md's Pull Requests](https://github.com/pytorch/executorch/blob/main/CONTRIBUTING.md#pull-requests) for ExecuTorch PR guidelines. [PLEASE REMOVE] If this PR closes an issue, please add a `Fixes #<issue-id>` line. [PLEASE REMOVE] If this PR introduces a fix or feature that should be the upcoming release notes, please add a "Release notes: <area>" label. For a list of available release notes labels, check out [CONTRIBUTING.md's Pull Requests](https://github.com/pytorch/executorch/blob/main/CONTRIBUTING.md#pull-requests). ### Test plan [PLEASE REMOVE] How did you test this PR? Please write down any manual commands you used and note down tests that you have written if applicable. (cherry picked from commit 8b11418)
Cherry picking #14427The cherry pick PR is at #14971 and it is recommended to link a critical cherry pick PR with an issue. The following tracker issues are updated: Details for Dev Infra teamRaised by workflow job |
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.
If the processed audio went through Mel transform, the spectrogram are float values. We should allow
Audioclass to be able to take this, since multimodal runner pybind API will have to be able to take processed input. Once we have the pybind API we can do something like:Summary
[PLEASE REMOVE] See CONTRIBUTING.md's Pull Requests for ExecuTorch PR guidelines.
[PLEASE REMOVE] If this PR closes an issue, please add a
Fixes #<issue-id>line.[PLEASE REMOVE] If this PR introduces a fix or feature that should be the upcoming release notes, please add a "Release notes: " label. For a list of available release notes labels, check out CONTRIBUTING.md's Pull Requests.
Test plan
[PLEASE REMOVE] How did you test this PR? Please write down any manual commands you used and note down tests that you have written if applicable.