Siddhant commited on
Commit
43f1cd1
1 Parent(s): 9bc6456

import from zenodo

Browse files
README.md ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - espnet
4
+ - audio
5
+ - automatic-speech-recognition
6
+ language: kr
7
+ datasets:
8
+ - zeroth_korean
9
+ license: cc-by-4.0
10
+ ---
11
+ ## Example ESPnet2 ASR model
12
+ ### `Hoon_Chung/zeroth_korean_asr_train_asr_transformer5_raw_bpe_valid.acc.ave`
13
+ ♻️ Imported from https://zenodo.org/record/4014588/
14
+
15
+ This model was trained by Hoon Chung using zeroth_korean/asr1 recipe in [espnet](https://github.com/espnet/espnet/).
16
+ ### Demo: How to use in ESPnet2
17
+ ```python
18
+ # coming soon
19
+ ```
20
+ ### Citing ESPnet
21
+ ```BibTex
22
+ @inproceedings{watanabe2018espnet,
23
+ author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson {Enrique Yalta Soplin} and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai},
24
+ title={{ESPnet}: End-to-End Speech Processing Toolkit},
25
+ year={2018},
26
+ booktitle={Proceedings of Interspeech},
27
+ pages={2207--2211},
28
+ doi={10.21437/Interspeech.2018-1456},
29
+ url={http://dx.doi.org/10.21437/Interspeech.2018-1456}
30
+ }
31
+ @inproceedings{hayashi2020espnet,
32
+ title={{Espnet-TTS}: Unified, reproducible, and integratable open source end-to-end text-to-speech toolkit},
33
+ author={Hayashi, Tomoki and Yamamoto, Ryuichi and Inoue, Katsuki and Yoshimura, Takenori and Watanabe, Shinji and Toda, Tomoki and Takeda, Kazuya and Zhang, Yu and Tan, Xu},
34
+ booktitle={Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
35
+ pages={7654--7658},
36
+ year={2020},
37
+ organization={IEEE}
38
+ }
39
+ ```
40
+ or arXiv:
41
+ ```bibtex
42
+ @misc{watanabe2018espnet,
43
+ title={ESPnet: End-to-End Speech Processing Toolkit},
44
+ author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Enrique Yalta Soplin and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai},
45
+ year={2018},
46
+ eprint={1804.00015},
47
+ archivePrefix={arXiv},
48
+ primaryClass={cs.CL}
49
+ }
50
+ ```
data/token_list/bpe_unigram5000/bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e1781ca9941f5115d6a5eb777e928060eb2adf20b87a166963a21e82399c88b
3
+ size 318981
exp/asr_stats_raw/train/feats_stats.npz ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/RESULTS.md ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <!-- Generated by scripts/utils/show_asr_result.sh -->
2
+ # RESULTS
3
+ ## Environments
4
+ - date: `Tue Sep 1 17:25:04 KST 2020`
5
+ - python version: `3.7.3 (default, Mar 27 2019, 22:11:17) [GCC 7.3.0]`
6
+ - espnet version: `espnet 0.7.0`
7
+ - pytorch version: `pytorch 1.2.0`
8
+ - Git hash: `df9b0b3aaf79bd7396b48b1d2c43e992466cb897`
9
+ - Commit date: `Tue Sep 1 16:55:23 2020 +0900`
10
+
11
+ ## asr_train_asr_transformer5_raw_bpe
12
+ ### WER
13
+
14
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
15
+ |---|---|---|---|---|---|---|---|---|
16
+ |inference_asr_model_valid.acc.ave/test_clean|457|9253|98.3|1.0|0.7|0.3|2.0|13.1|
17
+ |inference_asr_model_valid.acc.best/test_clean|457|9253|97.8|1.4|0.9|0.2|2.4|17.3|
18
+
19
+ ### CER
20
+
21
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
22
+ |---|---|---|---|---|---|---|---|---|
23
+ |inference_asr_model_valid.acc.ave/test_clean|457|28084|98.9|0.4|0.7|0.3|1.4|13.1|
24
+ |inference_asr_model_valid.acc.best/test_clean|457|28084|98.6|0.5|0.9|0.2|1.6|17.3|
25
+
26
+ ### TER
27
+
28
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
29
+ |---|---|---|---|---|---|---|---|---|
30
+ |inference_asr_model_valid.acc.ave/test_clean|457|14176|98.6|0.7|0.7|0.3|1.7|13.1|
31
+ |inference_asr_model_valid.acc.best/test_clean|457|14176|98.2|0.9|0.9|0.2|2.0|17.3|
32
+
exp/asr_train_asr_transformer5_raw_bpe/config.yaml ADDED
The diff for this file is too large to render. See raw diff
exp/asr_train_asr_transformer5_raw_bpe/images/acc.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/backward_time.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/cer.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/cer_ctc.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/forward_time.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/iter_time.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/loss.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/loss_att.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/loss_ctc.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/lr_0.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/optim_step_time.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/train_time.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/images/wer.png ADDED
Binary file
exp/asr_train_asr_transformer5_raw_bpe/valid.acc.ave_10best.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:860cd125d0df6e016ad9192c468c2b451d7a8ff8423aeb16c7a30838b5f68296
3
+ size 123796006
meta.yaml ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
1
+ espnet: 0.7.0
2
+ files:
3
+ asr_model_file: exp/asr_train_asr_transformer5_raw_bpe/valid.acc.ave_10best.pth
4
+ python: "3.7.3 (default, Mar 27 2019, 22:11:17) \n[GCC 7.3.0]"
5
+ timestamp: 1599197410.92735
6
+ torch: 1.2.0
7
+ yaml_files:
8
+ asr_train_config: exp/asr_train_asr_transformer5_raw_bpe/config.yaml