Whisper-live

What's Changed
* Backslash missing in example by t-nil in https://github.com/collabora/WhisperLive/pull/236
* Making backend arg safer by berkaybilik in https://github.com/collabora/WhisperLive/pull/243
* Pin silero VAD onnx model version to v4.0 by makaveli10 in https://github.com/collabora/WhisperLive/pull/247
* Fix deadlock issue in FFmpeg subprocess by ensuring stderr is consumed by sondt2709 in https://github.com/collabora/WhisperLive/pull/257
* Discard no speech segments by makaveli10 in https://github.com/collabora/WhisperLive/pull/262
* Initialize last_segment to None by makaveli10 in https://github.com/collabora/WhisperLive/pull/272
* Set compute_type based on device capability by makaveli10 in https://github.com/collabora/WhisperLive/pull/274

**Full Changelog**: https://github.com/collabora/WhisperLive/compare/v0.5.0...v0.5.1

What's Changed
* Improve cpu and gpu Dockerfiles, resulting in much smaller images by peldszus in https://github.com/collabora/WhisperLive/pull/206
* Add option: save network stream to local file while transcribing by fraic in https://github.com/collabora/WhisperLive/pull/192
* fix: limit CPU usage for VAD onnxruntime inference session by setting… by makaveli10 in https://github.com/collabora/WhisperLive/pull/215
* Add support for RTSP stream by dshepelev15 in https://github.com/collabora/WhisperLive/pull/212
* Make writing audio frames optional by makaveli10 in https://github.com/collabora/WhisperLive/pull/216
* Expose the srt file location of Transcription client by chien-liu in https://github.com/collabora/WhisperLive/pull/224
* Update tensorrt llm to v0.9.0 by makaveli10 in https://github.com/collabora/WhisperLive/pull/227
* Single model mode by peldszus in https://github.com/collabora/WhisperLive/pull/223

**Full Changelog**: https://github.com/collabora/WhisperLive/compare/v0.4.1...v0.5.0

What's Changed
* Don't clear server_error flag on close. by jsichi in https://github.com/collabora/WhisperLive/pull/187
* Fix client close by makaveli10 in https://github.com/collabora/WhisperLive/pull/190


**Full Changelog**: https://github.com/collabora/WhisperLive/compare/v0.4.0...v0.4.1

What's Changed
* build & push docker image on every new tag by makaveli10 in https://github.com/collabora/WhisperLive/pull/161
* Upgrade faster whisper==1.0.1 by makaveli10 in https://github.com/collabora/WhisperLive/pull/163
* Handle failure on systems without microphones by FlippFuzz in https://github.com/collabora/WhisperLive/pull/173
* Fix faster whisper version in setup.py by FlippFuzz in https://github.com/collabora/WhisperLive/pull/175
* Add support for processing same audio stream via multiple clients running different tasks. by jsichi in https://github.com/collabora/WhisperLive/pull/174

New Contributors
* FlippFuzz made their first contribution in https://github.com/collabora/WhisperLive/pull/173
* jsichi made their first contribution in https://github.com/collabora/WhisperLive/pull/174

**Full Changelog**: https://github.com/collabora/WhisperLive/compare/v0.3.0...v0.4.0

What's Changed
* Fix docker image gpu by makaveli10 in https://github.com/collabora/WhisperLive/pull/156
* fix: lanuguage, task prefix in decoder start ids by makaveli10 in https://github.com/collabora/WhisperLive/pull/157
* fix: cpu usage issue by makaveli10 in https://github.com/collabora/WhisperLive/pull/158


**Full Changelog**: https://github.com/collabora/WhisperLive/compare/v0.2.1...v0.3.0

- Add unittests
- update requirements to include `scipy`
- code format/quality check according to flake8
- `vad` is now an option that the client can set if they want to use it or not
- docker image for both cpu and gpu are pushed to ghcr on new tag