tonko22
tonko22
I followed setup instructions: clone github repo, pip install -r requirements.txt python verify.py wavs-test.txt phones-test.txt data.idx caused error: :~/projects/vosk$ python verify.py wavs-test.txt phones-test.txt data.idx Traceback (most recent call last): File...
Can you please provide standart datasets benchmarks (LibriSpeech/CHiME/VoxForge)? Also interested in accuracy scaling with decreasing dataset size.
I'm trying to use CTC WFST-search for rescoring with compiled TLG graph using this tutorial: [https://wenet-e2e.github.io/wenet/lm.html](url) and passing these parameters to decoder: ` config = { "model_path": f"wenet/{model_name}/final.zip", "dict_path": f"wenet/{model_name}/words.txt",...
Based on fluent analysis i see only one advantage: unimrcp requires installing and configuring, when asterisk-audiofork requires only the C-script to be dumbped called from dialplan. Is there more advantages....