README.md

metadata

language:
  - ur
license: apache-2.0
tags:
  - generated_from_trainer
  - hf-asr-leaderboard
  - robust-speech-event
datasets:
  - mozilla-foundation/common_voice_8_0
metrics:
  - wer
  - cer
base_model: facebook/wav2vec2-xls-r-300m
model-index:
  - name: wav2vec2-large-xls-r-300m-Urdu
    results:
      - task:
          type: automatic-speech-recognition
          name: Speech Recognition
        dataset:
          name: Common Voice 8
          type: mozilla-foundation/common_voice_8_0
          args: ur
        metrics:
          - type: wer
            value: 39.89
            name: Test WER
          - type: cer
            value: 16.7
            name: Test CER
new_version: kingabzpro/whisper-large-v3-turbo-urdu
pipeline_tag: automatic-speech-recognition

wav2vec2-large-xls-r-300m-Urdu

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

• Loss: 0.9889
• Wer: 0.5607
• Cer: 0.2370

Evaluation Commands

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

python eval.py --model_id kingabzpro/wav2vec2-large-xls-r-300m-Urdu --dataset mozilla-foundation/common_voice_8_0 --config ur --split test

Inference With LM

#pip install pyctcdecode kenlm

from datasets import load_dataset, Audio
from transformers import pipeline
model = "kingabzpro/wav2vec2-large-xls-r-300m-Urdu"
data = load_dataset("mozilla-foundation/common_voice_8_0",
                     "ur",
                     split="test", 
                     streaming=True, 
                     trust_remote_code=True)

sample_iter = iter(data.cast_column("audio", 
                    Audio(sampling_rate=16_000)))
sample = next(sample_iter)

asr = pipeline("automatic-speech-recognition", model=model)
prediction = asr(sample["audio"]["array"], 
                  chunk_length_s=5, 
                  stride_length_s=1)
prediction
# => {'text': 'مزدور تے کہ علاوہ سرکاری اور کاروباری لوگ ن ڈرپجے کام شروع کرتے'}

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 32
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
3.6398	30.77	400	3.3517	1.0	1.0
2.9225	61.54	800	2.5123	1.0	0.8310
1.2568	92.31	1200	0.9699	0.6273	0.2575
0.8974	123.08	1600	0.9715	0.5888	0.2457
0.7151	153.85	2000	0.9984	0.5588	0.2353
0.6416	184.62	2400	0.9889	0.5607	0.2370

Framework versions

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

Eval results on Common Voice 8 "test" (WER):

Without LM	With LM (run ./eval.py)
52.03	39.89