Emmanuel Schmidbauer

how to use custom trained checkpoint

1

Thanks for sharing this project! I've followed the instructions to train a custom model. The tensorboard is showing decent progress and audio predictions are starting to sound good. But I...

Trying out any whisper large (v2, v3, turbo) and it seems to not work, I get the same error regardless of model: ``` python import whisper_s2t model = whisper_s2t.load_model(model_identifier="large-v2", backend="hf",...

pretrained model (`enc_dec_fix_phonemes.pt`) for testing inference

3

This repo looks very interesting, thank you for sharing! I'd like to try inference out myself - is it possible for you to share a pretrained model? Or would you...

add support for `LiteWhisperConfig`

3

new whisper [model](https://huggingface.co/efficient-speech/lite-whisper-large-v3-turbo) uses configuration `LiteWhisperConfig` when i try to convert with `ct2-transformers-converter --model efficient-speech/lite-whisper-large-v3-turbo --output_dir lite-whisper-large-v3-turbo-ct2 --trust_remote_code` i get ```bash ValueError: No conversion is registered for the model configuration...

- Instantiate a fresh Silero VAD model per AudioProcessor instead of sharing the singleton from TranscriptionEngine, eliminating shared _state/_context across clients. - Keep VAC optional but ensure each connection now...

fix torch error

1

fixes error: ```shell torch/serialization.py", line 1470, in load raise pickle.UnpicklingError(_get_wo_message(str(e))) from None _pickle.UnpicklingError: Weights only load failed. This file can still be loaded, to do so you have two options,...

export `VibeVoice-Realtime-0.5B` to onnx for wasm

3

Would it be possible to export the `VibeVoice-Realtime-0.5B` model to onnx and be used with wasm?