File size: 1,223 Bytes
754e3f4
 
 
 
d3c8ec4
754e3f4
d3c8ec4
 
754e3f4
1b25e70
 
 
 
 
754e3f4
 
 
 
d3c8ec4
754e3f4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
language:
- en
tags:
- esb
datasets:
- esb/datasets
- kensho/spgispeech
---
To reproduce this run, first install Whisper from the Transformers compatible repo [patrickvonplaten/whisper](https://github.com/patrickvonplaten/whisper):
```
pip install git+https://github.com/openai/whisper.git
```
Then execute the command: 
```python 
#!/usr/bin/env bash
CUDA_VISIBLE_DEVICES=0 python run_speech_recognition_whisper.py \
	--model_name_or_path="medium.en" \
  --dataset_name="esb/datasets" \
  --dataset_config_name="spgispeech" \
	--max_steps="5000" \
	--output_dir="./" \
	--run_name="whisper-spgispeech" \
	--wandb_project="whisper" \
	--per_device_train_batch_size="64" \
	--per_device_eval_batch_size="16" \
	--logging_steps="25" \
	--learning_rate="1e-4" \
	--warmup_steps="500" \
	--report_to="wandb" \
	--preprocessing_num_workers="16" \
	--evaluation_strategy="steps" \
	--eval_steps="1000" \
	--save_strategy="steps" \
	--save_steps="1000" \
	--generation_max_length="224" \
	--length_column_name="input_lengths" \
	--gradient_checkpointing \
	--group_by_length \
	--freeze_encoder \
	--fp16 \
	--overwrite_output_dir \
	--do_train \
	--do_eval \
	--do_predict \
	--predict_with_generate \
	--use_auth_token

```