GigaSpeech ASR model
The following models are Chain models trained on GigaSpeech's Small (250 hours), Medium (1,000 hours), Large (2,500 hours) and Extra Large (10,000 hours) subsets respectively. See this page for their performance and their position in the GigaSpeech leaderboard.
GigaSpeech ASR S
- Date
- 2022-02-03
- Uploader
- uploaded by Yenda
- Recipe
- egs/gigaspeech/s5
- Kaldi Version
- f6f4cca
- Model Type
- Speech Recognition, Factored TDNN, LSTM, Chain
GigaSpeech ASR M
- Date
- 2022-02-03
- Uploader
- uploaded by Yenda
- Recipe
- egs/gigaspeech/s5
- Kaldi Version
- f6f4cca
- Model Type
- Speech Recognition, Factored TDNN, LSTM, Chain
GigaSpeech ASR L
- Date
- 2022-02-03
- Uploader
- uploaded by Yenda
- Recipe
- egs/gigaspeech/s5
- Kaldi Version
- f6f4cca
- Model Type
- Speech Recognition, Factored TDNN, LSTM, Chain
GigaSpeech ASR XL
- Date
- 2022-02-03
- Uploader
- uploaded by Yenda
- Recipe
- egs/gigaspeech/s5
- Kaldi Version
- f6f4cca
- Model Type
- Speech Recognition, Factored TDNN, LSTM, Chain
