Hugging Face's logo

comodoro
/
wav2vec2-xls-r-300m-west-slavic-cv8

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
hf-asr-leaderboard
mozilla-foundation/common_voice_8_0
robust-speech-event
xlsr-fine-tuning-week
Eval Results
Inference Endpoints
Model card Files Community
1
wav2vec2-xls-r-300m-west-slavic-cv8 / README.md
comodoro's picture
comodoro
Initial model
c6f57da
preview code
|
raw
history blame
4.21 kB
metadata
license: apache-2.0
tags:
  - generated_from_trainer
model-index:
  - name: wav2vec2-xls-r-300m-west-slavic-cv8
    results: []

wav2vec2-xls-r-300m-west-slavic-cv8

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 2.3462
  • Wer: 0.8556
  • Cer: 0.2799

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: 32
  • 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: 50
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer Cer
6.548 1.23 400 3.4763 1.0 1.0
3.42 2.45 800 3.3156 1.0 1.0
3.291 3.68 1200 3.2396 1.0 1.0
2.6515 4.91 1600 2.0422 0.9997 0.5835
1.7019 6.13 2000 1.6337 0.9893 0.4797
1.3604 7.36 2400 1.5221 0.9875 0.4463
1.1965 8.59 2800 1.5284 0.9766 0.4247
1.069 9.82 3200 1.5228 0.9672 0.4124
0.9536 11.04 3600 1.4059 0.9600 0.3868
0.8487 12.27 4000 1.4083 0.9501 0.3739
0.7655 13.5 4400 1.4079 0.9369 0.3612
0.6956 14.72 4800 1.4170 0.9411 0.3459
0.6287 15.95 5200 1.4000 0.9235 0.3384
0.561 17.18 5600 1.4735 0.9023 0.3295
0.5155 18.4 6000 1.5386 0.9202 0.3223
0.4864 19.63 6400 1.6186 0.9073 0.3259
0.4261 20.86 6800 1.6417 0.9217 0.3130
0.4051 22.09 7200 1.6295 0.8954 0.3026
0.3779 23.31 7600 1.8218 0.8979 0.3153
0.35 24.54 8000 1.7790 0.8921 0.3036
0.3343 25.77 8400 1.8588 0.9114 0.3072
0.3137 26.99 8800 1.8096 0.8756 0.2935
0.299 28.22 9200 1.9721 0.8863 0.3023
0.2894 29.45 9600 1.9907 0.8872 0.2958
0.2784 30.67 10000 1.9494 0.9090 0.2945
0.2662 31.9 10400 1.9952 0.8978 0.2935
0.2614 33.13 10800 2.0600 0.8949 0.2979
0.2401 34.36 11200 2.1180 0.8914 0.2950
0.2392 35.58 11600 2.1197 0.8713 0.2895
0.23 36.81 12000 2.1680 0.8713 0.2941
0.2246 38.04 12400 2.1526 0.8741 0.2879
0.2152 39.26 12800 2.2631 0.8790 0.2889
0.212 40.49 13200 2.2724 0.8661 0.2843
0.2044 41.72 13600 2.2438 0.8691 0.2878
0.2029 42.94 14000 2.2519 0.8577 0.2833
0.1972 44.17 14400 2.2697 0.8604 0.2813
0.1884 45.4 14800 2.3294 0.8662 0.2847
0.1877 46.63 15200 2.3077 0.8561 0.2793
0.1871 47.85 15600 2.3518 0.8563 0.2801
0.1838 49.08 16000 2.3462 0.8556 0.2799

Framework versions

  • Transformers 4.16.0.dev0
  • Pytorch 1.10.1+cu102
  • Datasets 1.18.3
  • Tokenizers 0.11.0