Update the PIP package

A tag to upload new PIP package.

Minor fixes

What's Changed

• Fix Golang example by @streamer45 in #496
• fix: rust example for v5 checkpoint by @rumbleFTW in #497
• VadIterator first chunk bag fx by @adamnsandle in #505
• Add java example for wav file & support V5 model by @yuguanqin in #506
• add csharp example by @nganju98 in #507
• downgrade onnxruntime dependency by @adamnsandle in #521
• код для тюнинга by @adamnsandle in #526
• add neg_threshold parameter explicitly by @adamnsandle in #528
• Adamnsandle by @adamnsandle in #529
• Fixed the pyaudio example can not run issue. by @gengyuchao in #539
• Update README.md by @adamnsandle in #540
• Update README.md by @adamnsandle in #541
• Update README.md by @snakers4 in #542
• Adamnsandle by @adamnsandle in #543
• Adamnsandle by @adamnsandle in #549

New Contributors

• @rumbleFTW made their first contribution in #497
• @yuguanqin made their first contribution in #506
• @nganju98 made their first contribution in #507
• @gengyuchao made their first contribution in #539

Full Changelog: v5.1...v5.1.1

v5.1

Experimental PIP package release

• Experimental pip-package release;
• Community PRs to update the examples;

What's Changed

• Adamnsandle by @adamnsandle in #481
• Update microphone_and_webRTC_integration.py by @eltociear in #475
• cpp example by @filtercodes in #482
• Update Golang example to support model v5 by @streamer45 in #489
• Create python-publish.yml by @adamnsandle in #492
• Adamnsandle by @adamnsandle in #493

New Contributors

• @eltociear made their first contribution in #475
• @filtercodes made their first contribution in #482

Full Changelog: v5.0...v5.1

# New V4 VAD Released

New V4 VAD Released

• Improved quality
• Improved perfomance
• Both 8k and 16k sampling rates are now supported by the ONNX model
• Batching is now supported by the ONNX model
• Added audio_forward method for one-line processing of a single or multiple audio without postprocessing
• Hotfix applied - wrong model was uploaded
• Minor hotfix re. PyTorch version

New V3 ONNX VAD Released

We finally were able to port a model to ONNX:

• Compact model (~100k params);
• Both PyTorch and ONNX models are not quantized;
• Same quality model as the latest best PyTorch release;
• Only 16kHz available now (ONNX has some issues with if-statements and / or tracing vs scripting) with cryptic errors;
• In our tests, on short audios (chunks) ONNX is 2-3x faster than PyTorch (this is mitigated with larger batches or long audios);
• Audio examples and non-core models moved out of the repo to save space;

Finally, V5 is here, 3x faster, supporting 6000+ languages!

Performance and Model Size

• 3x faster inference for TorchScript, 10% faster inference for ONNX;
• Now TorchScript is as fast as ONNX;
• Model size is 2x larger, 2MB vs. 1MB;

Quality

• The VAD supports more than 6,000 languages now;
• Significanly more robust on noisy data;
• Overall 5-7% quality increase on clean data;
• Quality difference for 8 kHz and 16 kHz is negligible now;
• Quality difference for different window sizes is negligible => window size was deprecated;
• Added benchmarks on 9 unique datasets (2 private) and one holistic multi-domain dataset;

Changes and deprecations

• ONNX opset 16;
• window_size_samples is deprecated - now the VAD only works with fixed size window;
• VAD now works with 8 kHz and 16 kHz sample rates, only with fixed 256 and 512 sample windows respectively;
• Slightly changed internal logic, now some context (part of previous chunk) is passed along with the current chunk;
• Sample rates that are a multiple of 16 kHz are still supported;

New V3 Silero VAD is Already Here

Main changes

• One VAD to rule them all! New model includes the functionality of the previous ones with improved quality and speed!
• Flexible sampling rate, 8000 Hz and 16000 Hz are supported;
• Flexible chunk size, minimum chunk size is just 30 milliseconds!
• 100k parameters;
• GPU and batching are supported;
• Radically simplified examples;

Migration

Please see the new examples.

New get_speech_timestamps is a simplified and unified version of the old deprecated get_speech_ts or get_speech_ts_adaptive methods.

speech_timestamps = get_speech_timestamps(wav, model, sampling_rate=16000)

New VADIterator class serves as an example for streaming tasks instead of old deprecated VADiterator and VADiteratorAdaptive.

vad_iterator = VADIterator(model)
window_size_samples = 1536

for i in range(0, len(wav), window_size_samples):
   speech_dict = vad_iterator(wav[i: i+ window_size_samples], return_seconds=True)
   if speech_dict:
       print(speech_dict, end=' ')
vad_iterator.reset_states()

V2 Legacy Release for History

This is a technical tag, so that users, who do now want to use newer models, could just checkout this tag.