ictnlp/StreamSpeech
StreamSpeech is an “All in One” seamless model for offline and simultaneous speech recognition, speech translation and speech synthesis.
PythonMIT
Issues
- 3
Low ASR-Bleu result with simul-s2st.sh
#12 opened by arararz - 12
- 1
RuntimeError: Input tensor has to be 2D. - When using Web GUI demo with own audio(.mp3)
#14 opened by Ztfrederickzheng - 0
- 5
- 1
some folders and files are missing
#7 opened by nattu22 - 1
pip install fairseq fails due to invalid metadata for PyYAML dependency in omegaconf
#11 opened by Gan-Xing - 1
Unable to run test script on local machine
#1 opened by maxjkwang - 0
- 0
- 5
Train on other language
#3 opened by yiwei0730 - 0
- 2
Error when loading speech_to_speech_ctc task
#2 opened by arararz