jonatasgrosman's picture
update eval
6cdc194
|
raw
history blame
2.29 kB
metadata
language:
  - en
license: apache-2.0
tags:
  - automatic-speech-recognition
  - mozilla-foundation/common_voice_8_0
  - en
  - robust-speech-event
datasets:
  - mozilla-foundation/common_voice_8_0
model-index:
  - name: XLS-R-1B - English
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 8
          type: mozilla-foundation/common_voice_8_0
          args: en
        metrics:
          - name: Test WER
            type: wer
            value: 21.05
          - name: Test CER
            type: cer
            value: 8.44
          - name: Test WER (+LM)
            type: wer
            value: 17.31
          - name: Test CER (+LM)
            type: cer
            value: 7.77
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Robust Speech Event - Dev Data
          type: speech-recognition-community-v2/dev_data
          args: en
        metrics:
          - name: Test WER
            type: wer
            value: 20.53
          - name: Test CER
            type: cer
            value: 9.31
          - name: Test WER (+LM)
            type: wer
            value: 17.7
          - name: Test CER (+LM)
            type: cer
            value: 8.93

XLS-R-1B-ENGLISH

Fine-tuned facebook/wav2vec2-xls-r-1b on English using the Common Voice 8. When using this model, make sure that your speech input is sampled at 16kHz.

This model has been fine-tuned thanks to the GPU credits generously given by the OVHcloud :)

The script used for training can be found here: https://github.com/jonatasgrosman/wav2vec2-sprint

Evaluation Commands

  1. To evaluate on mozilla-foundation/common_voice_8_0 with split test
python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-english --dataset mozilla-foundation/common_voice_8_0 --config en --split test
  1. To evaluate on speech-recognition-community-v2/dev_data
python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-english --dataset speech-recognition-community-v2/dev_data --config en --split validation --chunk_length_s 5.0 --stride_length_s 1.0