xlsr_hungarian_new / README.md
Akashpb13's picture
Update README.md
43b53a7
|
raw
history blame
3.13 kB
metadata
language:
  - hu
license: apache-2.0
tags:
  - automatic-speech-recognition
  - mozilla-foundation/common_voice_8_0
  - generated_from_trainer
  - hu
  - robust-speech-event
  - model_for_talk
datasets:
  - mozilla-foundation/common_voice_8_0
model-index:
  - name: Akashpb13/xlsr_hungarian_new
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice 8
          type: mozilla-foundation/common_voice_8_0
          args: hu
        metrics:
          - name: Test WER
            type: wer
            value: 0.2851621517163838
          - name: Test CER
            type: cer
            value: 0.06112982522287432
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Robust Speech Event - Dev Data
          type: speech-recognition-community-v2/dev_data
          args: hu
        metrics:
          - name: Test WER
            type: wer
            value: 0.2851621517163838
          - name: Test CER
            type: cer
            value: 0.06112982522287432

Akashpb13/xlsr_hungarian_new

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - hu dataset. It achieves the following results on evaluation set (which is 10 percent of train data set merged with invalidated data, reported, other and dev datasets):

  • Loss: 0.197464
  • Wer: 0.330094

Model description

"facebook/wav2vec2-xls-r-300m" was finetuned.

Intended uses & limitations

More information needed

Training and evaluation data

Training data - Common voice hungarian train.tsv, dev.tsv, invalidated.tsv, reported.tsv, other.tsv and validated.tsv Only those points were considered where upvotes were greater than downvotes and duplicates were removed after concatenation of all the datasets given in common voice 7.0

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: 0.000095637994662983496
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 13
  • gradient_accumulation_steps: 16
  • lr_scheduler_type: cosine_with_restarts
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 100
  • mixed_precision_training: Native AMP

Training results

Step Training Loss Validation Loss Wer
500 4.785300 0.952295 0.796236
1000 0.535800 0.217474 0.381613
1500 0.258400 0.205524 0.345056
2000 0.202800 0.198680 0.336264
2500 0.182700 0.197464 0.330094

Framework versions

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

Evaluation Commands

  1. To evaluate on mozilla-foundation/common_voice_8_0 with split test
python eval.py --model_id Akashpb13/xlsr_hungarian_new --dataset mozilla-foundation/common_voice_8_0 --config hu --split test