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

Update RESULTS.md

Browse Source
This commit is contained in:
hnluo 2023-07-21 17:16:50 +08:00 committed by GitHub
parent fdce4df431
commit 80ac3acb07
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 7 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
egs_modelscope/asr/paraformer/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch/RESULTS.md
View File

@ -103,12 +103,10 @@
- Lora info: lora_bias: "all", lora_list ['q','v'], lora_rank:8, lora_alpha:16, lora_dropout:0.1
- Decoding info: beam_size 1, average_num 10
| model | Training Data(h) | Trainable Parameters(M) | cn cer(%) | sichuan cer(%) |
|:-------------:|:----------------:|:-----------------------:|:---------:|:--------------:|
| Pretrain | | | 8.57 | 19.81 |
| | | | | |
| Full-tuning | 50 | 220.9 | 8.8 | 12 |
| Lora Finetune | 50 | 2.29 | 9.13 | 12.13 |
| | | | | |
| Full-tuning | 200 | 220.9 | 9.82 | 10.47 |
| Lora Finetune | 200 | 2.29 | 9.21 | 11.28 |
| model | Training Data(h) | Trainable Parameters(M) | Memory Usage(G) | common cer(%) | sichuan cer(%) |
|:---------------:|:------------------:|:-------------------------:|:-----------------:|:---------------:|:----------------:|
| Pretrain | | | | 8.57 | 19.81 |
| Full-tuning | 50 | 220.9 | 15 | 8.8 | 12 |
| Lora Finetune | 50 | 2.29 | 7 | 9.13 | 12.13 |
| Full-tuning | 200 | 220.9 | 15 | 9.82 | 10.47 |
| Lora Finetune | 200 | 2.29 | 7 | 9.21 | 11.28 |