This model is a fine-tuned version of facebook/wav2vec2-xls-r-1b on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - sah dataset. It achieves the following results on evaluation set (which is 10 percent of train data set merged with other and dev datasets):

Loss: 21.39
Wer: 30.99

Model description

"facebook/wav2vec2-xls-r-1b" was finetuned.

Intended uses & limitations

More information needed

Training and evaluation data

Training data - Common voice Finnish train.tsv, dev.tsv and other.tsv

Training procedure

For creating the train dataset, all possible datasets were appended and 90-10 split was used.

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.000045637994662983496
train_batch_size: 16
eval_batch_size: 16
seed: 13
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine_with_restarts
lr_scheduler_warmup_steps: 500
num_epochs: 15
mixed_precision_training: Native AMP

Training results

Step	Training Loss	Validation Loss	Wer
200	4.541600	1.044711	0.926395
400	1.013700	0.290368	0.401758
600	0.645000	0.232261	0.346555
800	0.467800	0.214120	0.318340
1000	0.502300	0.213995	0.309957

Framework versions

Transformers 4.16.0.dev0
Pytorch 1.10.0+cu102
Datasets 1.17.1.dev0
Tokenizers 0.10.3

Evaluation Commands

To evaluate on mozilla-foundation/common_voice_8_0 with split test

python eval.py --model_id sammy786/wav2vec2-xlsr-sakha --dataset mozilla-foundation/common_voice_8_0 --config sah --split test