File size: 2,596 Bytes
7247cd3
 
 
 
8d1bf65
 
7247cd3
 
 
 
 
 
 
 
 
 
 
 
8d1bf65
7247cd3
 
 
8d1bf65
7247cd3
 
 
8d1bf65
7247cd3
 
 
 
 
 
 
8d1bf65
7247cd3
8d1bf65
 
7247cd3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
---
license: apache-2.0
base_model: facebook/wav2vec2-large-lv60
tags:
- automatic-speech-recognition
- edinburghcstr/ami
- generated_from_trainer
datasets:
- ami
metrics:
- wer
model-index:
- name: facebook/wav2vec2-large-lv60
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: EDINBURGHCSTR/AMI - IHM
      type: ami
      config: ihm
      split: None
      args: 'Config: ihm, Training split: train, Eval split: validation'
    metrics:
    - name: Wer
      type: wer
      value: 0.9542044754234227
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# facebook/wav2vec2-large-lv60

This model is a fine-tuned version of [facebook/wav2vec2-large-lv60](https://huggingface.co/facebook/wav2vec2-large-lv60) on the EDINBURGHCSTR/AMI - IHM dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2723
- Wer: 0.9542

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 2.0
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step  | Validation Loss | Wer    |
|:-------------:|:------:|:-----:|:---------------:|:------:|
| 1.0919        | 0.1565 | 1000  | 1.0169          | 0.7064 |
| 1.4768        | 0.3131 | 2000  | 0.7156          | 0.4356 |
| 0.9728        | 0.4696 | 3000  | 0.6462          | 0.4030 |
| 0.5418        | 0.6262 | 4000  | 0.6171          | 0.3707 |
| 0.8492        | 0.7827 | 5000  | 0.5758          | 0.3695 |
| 1.4826        | 0.9393 | 6000  | 0.5801          | 0.3545 |
| 0.3274        | 1.0958 | 7000  | 0.5244          | 0.3375 |
| 0.2089        | 1.2523 | 8000  | 0.5047          | 0.3239 |
| 0.2916        | 1.4089 | 9000  | 0.4901          | 0.3171 |
| 0.1617        | 1.5654 | 10000 | 0.5070          | 0.3151 |
| 0.3815        | 1.7220 | 11000 | 0.4948          | 0.3180 |
| 1.0171        | 1.8785 | 12000 | 0.9465          | 0.8379 |


### Framework versions

- Transformers 4.42.0.dev0
- Pytorch 2.3.0a0+gitcd033a1
- Datasets 2.19.1
- Tokenizers 0.19.1