wav2vec2-classic-300m-norwegian-colab
Browse files
README.md
ADDED
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: facebook/wav2vec2-xls-r-300m
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- nb_samtale
|
8 |
+
metrics:
|
9 |
+
- wer
|
10 |
+
model-index:
|
11 |
+
- name: wav2vec2-classic-300m-norwegian-colab
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Automatic Speech Recognition
|
15 |
+
type: automatic-speech-recognition
|
16 |
+
dataset:
|
17 |
+
name: nb_samtale
|
18 |
+
type: nb_samtale
|
19 |
+
config: annotations
|
20 |
+
split: test
|
21 |
+
args: annotations
|
22 |
+
metrics:
|
23 |
+
- name: Wer
|
24 |
+
type: wer
|
25 |
+
value: 0.7528477035956058
|
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-classic-300m-norwegian-colab
|
32 |
+
|
33 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the nb_samtale dataset.
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 2.2190
|
36 |
+
- Wer: 0.7528
|
37 |
+
|
38 |
+
## Model description
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Intended uses & limitations
|
43 |
+
|
44 |
+
More information needed
|
45 |
+
|
46 |
+
## Training and evaluation data
|
47 |
+
|
48 |
+
More information needed
|
49 |
+
|
50 |
+
## Training procedure
|
51 |
+
|
52 |
+
### Training hyperparameters
|
53 |
+
|
54 |
+
The following hyperparameters were used during training:
|
55 |
+
- learning_rate: 0.0003
|
56 |
+
- train_batch_size: 16
|
57 |
+
- eval_batch_size: 8
|
58 |
+
- seed: 42
|
59 |
+
- gradient_accumulation_steps: 2
|
60 |
+
- total_train_batch_size: 32
|
61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
62 |
+
- lr_scheduler_type: linear
|
63 |
+
- lr_scheduler_warmup_steps: 500
|
64 |
+
- num_epochs: 30
|
65 |
+
- mixed_precision_training: Native AMP
|
66 |
+
|
67 |
+
### Training results
|
68 |
+
|
69 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
70 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
71 |
+
| 4.8141 | 2.57 | 400 | 3.0571 | 1.0 |
|
72 |
+
| 3.0777 | 5.14 | 800 | 2.9987 | 1.0 |
|
73 |
+
| 2.7311 | 7.72 | 1200 | 2.5705 | 0.9829 |
|
74 |
+
| 2.1302 | 10.29 | 1600 | 1.8399 | 0.9225 |
|
75 |
+
| 1.6827 | 12.86 | 2000 | 1.6372 | 0.8559 |
|
76 |
+
| 1.312 | 15.43 | 2400 | 1.8908 | 0.9172 |
|
77 |
+
| 0.9979 | 18.01 | 2800 | 1.7908 | 0.7890 |
|
78 |
+
| 0.7456 | 20.58 | 3200 | 1.8110 | 0.7720 |
|
79 |
+
| 0.592 | 23.15 | 3600 | 2.0024 | 0.7686 |
|
80 |
+
| 0.4946 | 25.72 | 4000 | 2.1173 | 0.7702 |
|
81 |
+
| 0.4093 | 28.3 | 4400 | 2.2190 | 0.7528 |
|
82 |
+
|
83 |
+
|
84 |
+
### Framework versions
|
85 |
+
|
86 |
+
- Transformers 4.35.2
|
87 |
+
- Pytorch 2.1.0+cu118
|
88 |
+
- Datasets 2.15.0
|
89 |
+
- Tokenizers 0.15.0
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1261955080
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:17fb026c80291e1a6f01e4182b81e9efc3dfd31f2c05649d7a0f16eb8c75815d
|
3 |
size 1261955080
|
runs/Nov25_14-51-21_acdb54445ae2/events.out.tfevents.1700924076.acdb54445ae2.1664.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2d445a9d35bb93734826836f5745bfeec34843bded69c00a85b0d6c7d9d7a6da
|
3 |
+
size 11555
|