metadata

language:
  - mn
license: apache-2.0
tags:
  - automatic-speech-recognition
  - generated_from_trainer
  - hf-asr-leaderboard
  - mn
  - model_for_talk
  - mozilla-foundation/common_voice_7_0
  - robust-speech-event
datasets:
  - mozilla-foundation/common_voice_7_0
model-index:
  - name: XLS-R-300M - Mongolian
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 7
          type: mozilla-foundation/common_voice_7_0
          args: mn
        metrics:
          - name: Test WER
            type: wer
            value: 44.709
          - name: Test CER
            type: cer
            value: 13.532
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Robust Speech Event - Dev Data
          type: speech-recognition-community-v2/dev_data
          args: mn
        metrics:
          - name: Test WER
            type: wer
            value: 76.643
          - name: Test CER
            type: cer
            value: 36.997
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Robust Speech Event - Test Data
          type: speech-recognition-community-v2/eval_data
          args: mn
        metrics:
          - name: Test WER
            type: wer
            value: 78.45

wav2vec2-large-xls-r-300m-mongolian

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - MN dataset. It achieves the following results on the evaluation set:

Loss: 0.6003
Wer: 0.4473

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: 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
1.3677	15.87	2000	0.6432	0.6198
1.1379	31.75	4000	0.6196	0.5592
1.0093	47.62	6000	0.5828	0.5117
0.8888	63.49	8000	0.5754	0.4822
0.7985	79.37	10000	0.5987	0.4690
0.697	95.24	12000	0.6014	0.4471

Framework versions

Transformers 4.16.0.dev0
Pytorch 1.10.1+cu102
Datasets 1.17.1.dev0
Tokenizers 0.11.0