speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
28719e181a
FunASR/egs/aishell/paraformerbert
History
游雁 28719e181a paraformer batch padding
2022-12-11 22:03:09 +08:00
..
conf update FunASR version==0.1.4 2022-12-09 22:16:23 +08:00
local update details 2022-11-29 14:26:55 +08:00
utils update FunASR version==0.1.4 2022-12-09 22:16:23 +08:00
path.sh update details 2022-11-29 14:26:55 +08:00
README.md update details 2022-11-29 14:26:55 +08:00
run.sh paraformer batch padding 2022-12-11 22:03:09 +08:00

README.md

ParaformerBert + specaug + speed perturbation + specaugmentation

Environments

  • date: Mon Nov 21 13:25:30 CST 2022
  • python version: 3.7.12
  • FunASR version: 0.1.0
  • pytorch version: pytorch 1.7.0

Config files

  • train config: conf/train_asr_paraformerbert_conformer_12e_6d_2048_256.yaml
  • model size: 46M
  • lm config: LM was not used
  • decode config: conf/decode_asr_transformer_noctc_1best.yaml (CTC was not used)

Results (CER)

testset CER(%)
dev 4.30
test 4.80