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
ebdf631d98
FunASR/egs/aishell/paraformer
History
嘉渊 ebdf631d98 update repo
2023-05-12 11:22:58 +08:00
..
conf update 2023-04-25 11:48:12 +08:00
local update 2023-04-25 17:44:14 +08:00
path.sh create 2022-11-26 21:56:51 +08:00
README.md create 2022-11-26 21:56:51 +08:00
run.sh update repo 2023-05-12 11:22:58 +08:00
utils modify aishell transformer directory name 2023-02-01 17:47:44 +08:00

README.md

Paraformer

pretrained model in ModelScope：speech_paraformer_asr_nat-aishell1-pytorch

Training Config

  • Feature info: using 80 dims fbank, global cmvn, speed perturb(0.9, 1.0, 1.1), specaugment
  • Train info: lr 5e-4, batch_size 25000, 2 gpu(Tesla V100), acc_grad 1, 50 epochs
  • Train config: conf/train_asr_paraformer_conformer_12e_6d_2048_256.yaml
  • LM config: LM was not used

Results (CER)

  • Decode config: conf/decode_asr_transformer_noctc_1best.yaml (ctc weight:0.0)
testset CER(%)
dev 4.66
test 5.11
  • Decode config: conf/decode_asr_transformer.yaml (ctc weight:0.5)
testset CER(%)
dev 4.52
test 4.94