wav2vec2-large-mms-1b-urmi-christian-nostress
Browse files- README.md +82 -0
- adapter.aii.safetensors +3 -0
- preprocessor_config.json +9 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
+
base_model: facebook/mms-1b-all
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- nena_speech_1_0_test
|
8 |
+
metrics:
|
9 |
+
- wer
|
10 |
+
model-index:
|
11 |
+
- name: wav2vec2-large-mms-1b-urmi-christian-nostress
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Automatic Speech Recognition
|
15 |
+
type: automatic-speech-recognition
|
16 |
+
dataset:
|
17 |
+
name: nena_speech_1_0_test
|
18 |
+
type: nena_speech_1_0_test
|
19 |
+
config: urmi (christian)
|
20 |
+
split: test
|
21 |
+
args: urmi (christian)
|
22 |
+
metrics:
|
23 |
+
- name: Wer
|
24 |
+
type: wer
|
25 |
+
value: 0.944954128440367
|
26 |
+
---
|
27 |
+
|
28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
+
should probably proofread and complete it, then remove this comment. -->
|
30 |
+
|
31 |
+
# wav2vec2-large-mms-1b-urmi-christian-nostress
|
32 |
+
|
33 |
+
This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the nena_speech_1_0_test dataset.
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.6617
|
36 |
+
- Wer: 0.9450
|
37 |
+
- Cer: 0.1846
|
38 |
+
|
39 |
+
## Model description
|
40 |
+
|
41 |
+
More information needed
|
42 |
+
|
43 |
+
## Intended uses & limitations
|
44 |
+
|
45 |
+
More information needed
|
46 |
+
|
47 |
+
## Training and evaluation data
|
48 |
+
|
49 |
+
More information needed
|
50 |
+
|
51 |
+
## Training procedure
|
52 |
+
|
53 |
+
### Training hyperparameters
|
54 |
+
|
55 |
+
The following hyperparameters were used during training:
|
56 |
+
- learning_rate: 0.001
|
57 |
+
- train_batch_size: 32
|
58 |
+
- eval_batch_size: 8
|
59 |
+
- seed: 42
|
60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
+
- lr_scheduler_type: linear
|
62 |
+
- lr_scheduler_warmup_steps: 100
|
63 |
+
- num_epochs: 5
|
64 |
+
|
65 |
+
### Training results
|
66 |
+
|
67 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
68 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
|
69 |
+
| 14.5154 | 0.78 | 25 | 10.7711 | 1.0 | 0.9743 |
|
70 |
+
| 7.3688 | 1.56 | 50 | 6.3951 | 1.0 | 0.9408 |
|
71 |
+
| 3.895 | 2.34 | 75 | 2.4755 | 1.0 | 0.7579 |
|
72 |
+
| 1.6699 | 3.12 | 100 | 0.8689 | 0.9725 | 0.2286 |
|
73 |
+
| 1.1293 | 3.91 | 125 | 0.7194 | 0.9633 | 0.1981 |
|
74 |
+
| 0.9637 | 4.69 | 150 | 0.6617 | 0.9450 | 0.1846 |
|
75 |
+
|
76 |
+
|
77 |
+
### Framework versions
|
78 |
+
|
79 |
+
- Transformers 4.34.1
|
80 |
+
- Pytorch 2.1.0+cu121
|
81 |
+
- Datasets 2.14.6
|
82 |
+
- Tokenizers 0.14.1
|
adapter.aii.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bf4701fa1c7ef165250b34603aacf66ab04c81c7d6e60b68ec93d06ff4cb9ccf
|
3 |
+
size 8839532
|
preprocessor_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
4 |
+
"feature_size": 1,
|
5 |
+
"padding_side": "right",
|
6 |
+
"padding_value": 0.0,
|
7 |
+
"return_attention_mask": true,
|
8 |
+
"sampling_rate": 16000
|
9 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5244d1b3bd6dc3cf8086690ed2663ba1acf5f3d260b75ff87b1e7522849c89f8
|
3 |
+
size 4600
|