来自cslt Wiki
(差异) ←上一版本 | 最后版本 (差异) | 下一版本→ (差异)
RNN AM
- AM training on big dataset --mengyuan
- Further test nnet3-ctc training on 4000h 8k & 10000h 16k dataset. Tune decoding configures (phone-lm-weight, acwt, blank-scale), but ctc is still about 5 percent worse than standard nnet3.
