Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,53 @@
|
|
1 |
---
|
2 |
license: cc-by-nc-4.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: cc-by-nc-4.0
|
3 |
+
language:
|
4 |
+
- th
|
5 |
+
metrics:
|
6 |
+
- cer
|
7 |
+
- wer
|
8 |
+
library_name: espnet
|
9 |
+
pipeline_tag: automatic-speech-recognition
|
10 |
---
|
11 |
+
# Model Card for Model ID
|
12 |
+
|
13 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
14 |
+
|
15 |
+
This is the baseline model of Khummuang in [Thai-dialect corpus](https://github.com/SLSCU/thai-dialect-corpus).
|
16 |
+
|
17 |
+
The training recipe was based on wsj recipe in [espnet](https://github.com/espnet/espnet/).
|
18 |
+
|
19 |
+
|
20 |
+
|
21 |
+
### Model Description
|
22 |
+
|
23 |
+
<!-- Provide a longer summary of what this model is. -->
|
24 |
+
|
25 |
+
This model is Hybrid CTC/Attention model with pre-trained HuBERT encoder.
|
26 |
+
|
27 |
+
|
28 |
+
|
29 |
+
## Evaluation
|
30 |
+
|
31 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
32 |
+
|
33 |
+
For evaluation, the metrics are CER and WER. before WER evaluation, transcriptions were re-tokenized using [PyThaiNLP](https://github.com/PyThaiNLP/pythainlp)
|
34 |
+
|
35 |
+
|Micro CER|Macro CER|Survival CER|E-commerce WER|Micro WER|Macro WER|Survival WER|E-commerce WER|
|
36 |
+
|---|---|---|---|---|---|---|---|
|
37 |
+
|5.35|5.65|6.29|5.02|7.53|8.73|11.38|6.09|
|
38 |
+
<!-- |8.08|11.51|17.39|5.63|12.18|16.65|25.58|7.72| -->
|
39 |
+
<!-- |18.17|22.38|31.01|13.75|31.74|37.68|50.54|24.82| -->
|
40 |
+
|
41 |
+
## Paper
|
42 |
+
|
43 |
+
[Thai Dialect Corpus and Transfer-based Curriculum Learning Investigation for Dialect Automatic Speech Recognition](https://www.isca-speech.org/archive/pdfs/interspeech_2023/suwanbandit23_interspeech.pdf)
|
44 |
+
```
|
45 |
+
@inproceedings{suwanbandit23_interspeech,
|
46 |
+
author={Artit Suwanbandit and Burin Naowarat and Orathai Sangpetch and Ekapol Chuangsuwanich},
|
47 |
+
title={{Thai Dialect Corpus and Transfer-based Curriculum Learning Investigation for Dialect Automatic Speech Recognition}},
|
48 |
+
year=2023,
|
49 |
+
booktitle={Proc. INTERSPEECH 2023},
|
50 |
+
pages={4069--4073},
|
51 |
+
doi={10.21437/Interspeech.2023-1828}
|
52 |
+
}
|
53 |
+
```
|