[RNE Rewrite] feat: add Whisper STT pipeline#1302
Open
barhanc wants to merge 18 commits into
Open
Conversation
barhanc
force-pushed
the
@bh/stt
branch
2 times, most recently
from
July 13, 2026 14:19
809853f to
6bfd3ad
Compare
barhanc
force-pushed
the
@bh/stt
branch
2 times, most recently
from
July 17, 2026 13:02
f1d6450 to
adb83a9
Compare
barhanc
marked this pull request as ready for review
July 17, 2026 13:37
msluszniak
reviewed
Jul 17, 2026
msluszniak
reviewed
Jul 17, 2026
- Remove unused react-native-fs dep from speech example app - Rename WhisperSttModel.fsmnVoiceActivityDetectorPath (string) to fsmnVadModel (FsmnVadModel) so defaultOptions flow through correctly - Update useSpeechToText hook to read model path from fsmnVadModel and reconstruct the object with the local path after download - Update all Whisper entries in models.ts accordingly; drop the now redundant FSMN_VAD_XNNPACK_FP32_URL constant - Replace voiceDetector.push() with detectVoiceOnStream() to match the renamed streaming API
Add ModelPicker component (matching CV app's horizontal chip scroller) and replace the inline fixed chip row on the STT screen with it.
- stt/ → microphone-transcription/ - stt-audio/ → audio-file-transcription/ Updated route names in _layout.tsx and navigation paths in index.tsx accordingly.
…port - Replace bhanc/scratch URLs with official software-mansion HuggingFace repos - Use BASE_URL/NEXT_VERSION_TAG pattern consistent with other models - Add MLX bf16 model configs for all Whisper variants (tiny/base/small, en/multilingual) - Add MLX model options to both example screens (iOS only)
Member
Author
|
@msluszniak I've updated the models on HF and added the MLX variant. I can't unfortunately test them since I don't have iOS device available. |
Member
|
@barhanc Ok, I will test mlx delegates tomorrow |
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
2 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.
Description
Adds a full Whisper speech-to-text task pipeline, React hook, model registry entries, and an interactive example screen.
Introduces a breaking change?
Type of change
Tested on
Testing instructions
Screenshots
Related issues
Closes #1208
Checklist
Additional notes
Needs to wait for #1309 before merging so that the paths in
models.tscan be pointed at SWM HF.