Skip to content

Tags: clamsproject/app-whisper-wrapper

Tags

v12

added support for arm64 container image

v11

- Now based on whisper 240930 version with support of `turbo` model

- Beam search size is set to 5 for the decoder, following the `whisper` command's default
- (temporarily) Disabled multiprocessing web app via gunicorn and fell back to flask built-in to work around CUDA memory issue

v10

This version adds some delegation parameters to whisper.transcribe

- `task`: delegate to `--task`
- `initialPrompt`: delegate to `--initial-prompt`
- `conditionOnPreviousText`: delegate to `--condition-on-previous-text`
- `noSpeechThreshold`: delegate to `--no-speech-threshold`

v9

fixed inability to handle concurrent requests

v8

SDK version update

v7

- Fixed minor typos in parameters

- Updated to clams-python 1.2.4, and added `cli.py` interface

v6

v6 fixes a bug running en-only large models since there is no en-only…

… large model

v5

This version include addition of `modelLang` param and many other cha…

…nges

* **(BIG CHANGE)** timeunit is now millisecond (integer)
* `modelLang` parameter is added for selecting language to instruct Whisper to use, Whisper will run in language detection mode if this parameter is not given.
* When `modelLang` is set to `en`, load english only models instead of multilingual ones for speed and performance
* skips empty segments from whisper when generating output MMIF
* updated to clams-python 1.1.1

v4

* updated to the latest whisper models

* now ouputs `Sentence` annotations based on whisper's segmentation

v3

updated to sdk 1.0.9, added model picker as runtime param