中文Sign up

DrishtiSharma/wav2vec2-xls-r-300m-mt-o1

🤗 Hugging Face sourceautomatic-speech-recognitionapache-2.08.8 GBother✓ 2 checksumsupdated today
Submit in one command

Run it next to your model folder. It makes the torrent, checks your files against Hugging Face, and submits it. You just start seeding and paste your key from your account. It only reads your files and never changes them. Read the script first if you like.

curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo DrishtiSharma/wav2vec2-xls-r-300m-mt-o1 ./model-folder
Needs a seeder →

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - MT dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1987
  • Wer: 0.1920

Evaluation Commands

  1. To evaluate on mozilla-foundation/common_voice_8_0 with test split

python eval.py --model_id DrishtiSharma/wav2vec2-xls-r-300m-mt-o1 --dataset mozilla-foundation/common_voice_8_0 --config mt --split test --log_outputs

  1. To evaluate on speech-recognition-community-v2/dev_data

Maltese language not found in speech-recognition-community-v2/dev_data!

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 7e-05
  • train_batch_size: 32
  • eval_batch_size: 1
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 2000
  • num_epochs: 100.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
1.1721 18.02 2000 0.3831 0.4066
0.7849 36.04 4000 0.2191 0.2417
0.6723 54.05 6000 0.2056 0.2134
0.6015 72.07 8000 0.2008 0.2031
0.5386 90.09 10000 0.1967 0.1953

Framework versions

  • Transformers 4.17.0.dev0
  • Pytorch 1.10.2+cu102
  • Datasets 1.18.2.dev0
  • Tokenizers 0.11.0