File size: 2,246 Bytes
b99a621
0a1c33d
b6994be
0a1c33d
 
 
b99a621
09dc80f
f567303
b324d03
 
86f38f9
ef8c3fa
38706e1
86f38f9
38706e1
 
3a12e41
 
b99a621
ba980b2
3a12e41
0a1c33d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a050237
a90a966
0a1c33d
 
 
a050237
 
0a1c33d
 
 
674aa77
a050237
0a1c33d
 
 
3a12e41
 
0a1c33d
 
3a12e41
b99a621
86f38f9
b324d03
b99a621
86f38f9
 
 
 
 
b99a621
 
86f38f9
 
 
b6994be
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
python run_speech_recognition_ctc.py \
	--dataset_name="marinone94/nst_sv" \
	--model_name_or_path="KBLab/wav2vec2-large-voxrex" \
	--dataset_config_name="distant_channel" \
	--train_split_name="train" \
	--eval_split_name="test" \
	--output_dir="./" \
	--overwrite_output_dir \
	--num_train_epochs="2" \
	--per_device_train_batch_size="32" \
	--per_device_eval_batch_size="32" \
	--gradient_accumulation_steps="4" \
	--learning_rate="7.5e-4" \
	--warmup_ratio="0.02" \
	--length_column_name="input_length" \
	--evaluation_strategy="steps" \
	--save_strategy="steps" \
	--eval_steps="100" \
	--save_steps="100" \
	--text_column_name="sentence" \
	--chars_to_ignore , ? . ! \- \; \: \" β€œ % β€˜ ” οΏ½ β€” ’ … – \
	--logging_steps="20" \
	--layerdrop="0.0" \
	--activation_dropout="0.1" \
	--save_total_limit="2" \
	--freeze_feature_encoder \
	--feat_proj_dropout="0.0" \
	--mask_time_prob="0.75" \
	--mask_time_length="10" \
	--mask_feature_prob="0.25" \
	--mask_feature_length="64" \
	--gradient_checkpointing \
	--use_auth_token \
	--fp16 \
	--group_by_length \
	--do_train --do_eval \
	--push_to_hub

python run_speech_recognition_ctc.py \
	--dataset_name="mozilla-foundation/common_voice_8_0" \
	--model_name_or_path="marinone94/xls-r-300m-sv-robust" \
	--dataset_config_name="sv-SE" \
	--train_split_name="train+validation" \
	--eval_split_name="test" \
	--output_dir="./" \
	--overwrite_output_dir \
	--num_train_epochs="50" \
	--per_device_train_batch_size="32" \
	--per_device_eval_batch_size="32" \
	--gradient_accumulation_steps="4" \
	--learning_rate="2.5e-4" \
	--warmup_steps="2000" \
	--length_column_name="input_length" \
	--evaluation_strategy="steps" \
	--save_strategy="steps" \
	--eval_steps="100" \
	--save_steps="100" \
	--text_column_name="sentence" \
	--chars_to_ignore , ? . ! \- \; \: \" β€œ % β€˜ ” οΏ½ β€” ’ … – \
	--logging_steps="20" \
	--layerdrop="0.0" \
	--activation_dropout="0.1" \
	--save_total_limit="2" \
	--freeze_feature_encoder \
	--feat_proj_dropout="0.0" \
	--mask_time_prob="0.75" \
	--mask_time_length="10" \
	--mask_feature_prob="0.25" \
	--mask_feature_length="64" \
	--gradient_checkpointing \
	--use_auth_token \
	--fp16 \
	--group_by_length \
	--do_train --do_eval \
	--push_to_hub