Commit
•
68cba10
1
Parent(s):
7aa917b
Update README.md
Browse files
README.md
CHANGED
@@ -1,12 +1,32 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
tags:
|
4 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
metrics:
|
6 |
-
- wer
|
7 |
model-index:
|
8 |
-
- name: whisper-small-mn-7
|
9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
---
|
11 |
|
12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -47,6 +67,21 @@ The following hyperparameters were used during training:
|
|
47 |
- training_steps: 15000
|
48 |
- mixed_precision_training: Native AMP
|
49 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
50 |
### Training results
|
51 |
|
52 |
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
tags:
|
4 |
+
- whisper-event
|
5 |
+
- hf-asr-leaderboard
|
6 |
+
- generated_from_trainer
|
7 |
+
datasets:
|
8 |
+
- mozilla-foundation/common_voice_11_0
|
9 |
+
- google/fleurs
|
10 |
+
- bayartsogt/ulaanbal-v0
|
11 |
metrics:
|
12 |
+
- wer
|
13 |
model-index:
|
14 |
+
- name: whisper-small-mn-7-bayartsogt
|
15 |
+
results:
|
16 |
+
- task:
|
17 |
+
name: Automatic Speech Recognition
|
18 |
+
type: automatic-speech-recognition
|
19 |
+
dataset:
|
20 |
+
name: Common Voice 11.0
|
21 |
+
type: mozilla-foundation/common_voice_11_0
|
22 |
+
config: mn
|
23 |
+
split: test
|
24 |
+
args:
|
25 |
+
language: mn
|
26 |
+
metrics:
|
27 |
+
- name: Wer
|
28 |
+
type: wer
|
29 |
+
value: 32.64693030369237
|
30 |
---
|
31 |
|
32 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
67 |
- training_steps: 15000
|
68 |
- mixed_precision_training: Native AMP
|
69 |
|
70 |
+
### Training script
|
71 |
+
```bash
|
72 |
+
python train.py \
|
73 |
+
--train_datasets "mozilla-foundation/common_voice_11_0|mn|train+validation,google/fleurs|mn_mn|train+validation,bayartsogt/ulaanbal-v0||train" \
|
74 |
+
--eval_datasets "mozilla-foundation/common_voice_11_0|mn|test" \
|
75 |
+
--whisper-size "small" \
|
76 |
+
--language "mn,Mongolian" \
|
77 |
+
--keep-chars " абвгдеёжзийклмноөпрстуүфхцчшъыьэюя.,?!" \
|
78 |
+
--train-batch-size 32 \
|
79 |
+
--eval-batch-size 32 \
|
80 |
+
--max-steps 15000 \
|
81 |
+
--num-workers 8 \
|
82 |
+
--version 7 \
|
83 |
+
```
|
84 |
+
|
85 |
### Training results
|
86 |
|
87 |
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|