File size: 1,802 Bytes
63b88a2
 
 
 
 
 
 
bcd2295
63b88a2
 
 
 
7bf3b4d
 
 
 
 
 
 
 
 
 
 
 
 
 
63b88a2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
language:
- bg
license: apache-2.0
tags:
- whisper-event
- generated_from_trainer
- hf-asr-leaderboard
datasets:
- mozilla-foundation/common_voice_11_0
model-index:
- name: Whisper Medium Bulgarian
  results: 
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: Common Voice 11.0
      type: mozilla-foundation/common_voice_11_0
      config: bg
      split: test
      args: bg
    metrics:
    - name: Wer
      type: wer
      value: 18.7606
---

<!-- 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. -->

# Whisper Medium Bulgarian

This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on the Common Voice 11.0 dataset.
It achieves the following results on the evaluation set:
- eval_loss: 0.3423
- eval_wer: 18.7606
- eval_runtime: 1297.1096
- eval_samples_per_second: 1.402
- eval_steps_per_second: 0.044
- epoch: 5.48
- step: 400

## 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: 1e-05
- train_batch_size: 16
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 1000
- mixed_precision_training: Native AMP

### Framework versions

- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2