Update README.md
Browse files
README.md
CHANGED
@@ -8,7 +8,7 @@ tags:
|
|
8 |
datasets:
|
9 |
- common_voice
|
10 |
model-index:
|
11 |
-
- name: Wav2Vec2 XLS-R 300M Cantonese (zh-HK)
|
12 |
results:
|
13 |
- task:
|
14 |
name: Automatic Speech Recognition
|
@@ -20,7 +20,7 @@ model-index:
|
|
20 |
metrics:
|
21 |
- name: Test CER
|
22 |
type: cer
|
23 |
-
value:
|
24 |
- task:
|
25 |
name: Automatic Speech Recognition
|
26 |
type: automatic-speech-recognition
|
@@ -31,45 +31,34 @@ model-index:
|
|
31 |
metrics:
|
32 |
- name: Test CER
|
33 |
type: cer
|
34 |
-
value: 56.
|
35 |
---
|
36 |
|
37 |
-
# Wav2Vec2 XLS-R 300M Cantonese (zh-HK)
|
38 |
|
39 |
-
Wav2Vec2 XLS-R 300M Cantonese (zh-HK)
|
40 |
|
41 |
This model was trained using HuggingFace's PyTorch framework and is part of the [Robust Speech Challenge Event](https://discuss.huggingface.co/t/open-to-the-community-robust-speech-recognition-challenge/13614) organized by HuggingFace. All training was done on a Tesla V100, sponsored by OVH.
|
42 |
|
43 |
-
All necessary scripts used for training could be found in the [Files and versions](https://huggingface.co/w11wo/wav2vec2-xls-r-300m-zh-HK-
|
44 |
-
|
45 |
-
As for the N-gram language model training, we followed the [blog post tutorial](https://huggingface.co/blog/wav2vec2-with-ngram) provided by HuggingFace.
|
46 |
|
47 |
## Model
|
48 |
|
49 |
-
| Model
|
50 |
-
|
|
51 |
-
| `wav2vec2-xls-r-300m-zh-HK-
|
52 |
|
53 |
## Evaluation Results
|
54 |
|
55 |
-
The model achieves the following results on evaluation
|
56 |
-
|
57 |
-
| Dataset | CER |
|
58 |
-
| -------------------------------- | ------ |
|
59 |
-
| `Common Voice` | 31.73% |
|
60 |
-
| `Robust Speech Event - Dev Data` | 56.60% |
|
61 |
|
62 |
-
|
63 |
-
|
64 |
-
|
|
65 |
-
|
|
66 |
-
| `Common Voice` | 12.14% |
|
67 |
-
| `Robust Speech Event - Dev Data` | 56.86% |
|
68 |
|
69 |
## Training procedure
|
70 |
|
71 |
-
The training process did not involve the addition of a language model. The following results were simply lifted from the original automatic speech recognition [model training](https://huggingface.co/w11wo/wav2vec2-xls-r-300m-korean).
|
72 |
-
|
73 |
### Training hyperparameters
|
74 |
|
75 |
The following hyperparameters were used during training:
|
@@ -171,7 +160,7 @@ Do consider the biases which came from pre-training datasets that may be carried
|
|
171 |
|
172 |
## Authors
|
173 |
|
174 |
-
Wav2Vec2 XLS-R 300M Cantonese (zh-HK)
|
175 |
|
176 |
## Framework versions
|
177 |
|
|
|
8 |
datasets:
|
9 |
- common_voice
|
10 |
model-index:
|
11 |
+
- name: Wav2Vec2 XLS-R 300M Cantonese (zh-HK)
|
12 |
results:
|
13 |
- task:
|
14 |
name: Automatic Speech Recognition
|
|
|
20 |
metrics:
|
21 |
- name: Test CER
|
22 |
type: cer
|
23 |
+
value: 31.73
|
24 |
- task:
|
25 |
name: Automatic Speech Recognition
|
26 |
type: automatic-speech-recognition
|
|
|
31 |
metrics:
|
32 |
- name: Test CER
|
33 |
type: cer
|
34 |
+
value: 56.60
|
35 |
---
|
36 |
|
37 |
+
# Wav2Vec2 XLS-R 300M Cantonese (zh-HK)
|
38 |
|
39 |
+
Wav2Vec2 XLS-R 300M Cantonese (zh-HK) is an automatic speech recognition model based on the [XLS-R](https://arxiv.org/abs/2111.09296) architecture. This model is a fine-tuned version of [Wav2Vec2-XLS-R-300M](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the `zh-HK` subset of the [Common Voice](https://huggingface.co/datasets/common_voice) dataset.
|
40 |
|
41 |
This model was trained using HuggingFace's PyTorch framework and is part of the [Robust Speech Challenge Event](https://discuss.huggingface.co/t/open-to-the-community-robust-speech-recognition-challenge/13614) organized by HuggingFace. All training was done on a Tesla V100, sponsored by OVH.
|
42 |
|
43 |
+
All necessary scripts used for training could be found in the [Files and versions](https://huggingface.co/w11wo/wav2vec2-xls-r-300m-zh-HK-v2/tree/main) tab, as well as the [Training metrics](https://huggingface.co/w11wo/wav2vec2-xls-r-300m-zh-HK-v2/tensorboard) logged via Tensorboard.
|
|
|
|
|
44 |
|
45 |
## Model
|
46 |
|
47 |
+
| Model | #params | Arch. | Training/Validation data (text) |
|
48 |
+
| ------------------------------ | ------- | ----- | ------------------------------- |
|
49 |
+
| `wav2vec2-xls-r-300m-zh-HK-v2` | 300M | XLS-R | `Common Voice zh-HK` Dataset |
|
50 |
|
51 |
## Evaluation Results
|
52 |
|
53 |
+
The model achieves the following results on evaluation:
|
|
|
|
|
|
|
|
|
|
|
54 |
|
55 |
+
| Dataset | Loss | CER |
|
56 |
+
| -------------------------------- | ------ | ------ |
|
57 |
+
| `Common Voice` | 0.8089 | 31.73% |
|
58 |
+
| `Robust Speech Event - Dev Data` | N/A | 56.60% |
|
|
|
|
|
59 |
|
60 |
## Training procedure
|
61 |
|
|
|
|
|
62 |
### Training hyperparameters
|
63 |
|
64 |
The following hyperparameters were used during training:
|
|
|
160 |
|
161 |
## Authors
|
162 |
|
163 |
+
Wav2Vec2 XLS-R 300M Cantonese (zh-HK) was trained and evaluated by [Wilson Wongso](https://w11wo.github.io/). All computation and development are done on OVH Cloud.
|
164 |
|
165 |
## Framework versions
|
166 |
|