metadata

language:
  - zh
license: apache-2.0
datasets:
  - mozilla-foundation/common_voice_16_0
model-index:
  - name: Wav2Vec2-BERT - Alvin
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: mozilla-foundation/common_voice_16_0 yue
          type: mozilla-foundation/common_voice_16_0
          config: yue
          split: test
          args: yue
        metrics:
          - name: Normalized CER
            type: cer
            value: 20.5

Wav2Vec2-BERT - Alvin

This model is a fine-tuned version of facebook/w2v-bert-2.0. This has a CER of 20.5

Training and evaluation data

For training, three datasets were used:

Common Voice 16 zh-HK and yue Train Set
CantoMap: Winterstein, Grégoire, Tang, Carmen and Lai, Regine (2020) "CantoMap: a Hong Kong Cantonese MapTask Corpus", in Proceedings of The 12th Language Resources and Evaluation Conference, Marseille: European Language Resources Association, p. 2899-2906.
Cantonse-ASR: Yu, Tiezheng, Frieske, Rita, Xu, Peng, Cahyawijaya, Samuel, Yiu, Cheuk Tung, Lovenia, Holy, Dai, Wenliang, Barezi, Elham, Chen, Qifeng, Ma, Xiaojuan, Shi, Bertram, Fung, Pascale (2022) "Automatic Speech Recognition Datasets in Cantonese: A Survey and New Dataset", 2022. Link: https://arxiv.org/pdf/2201.02419.pdf

Code Example

from transformers import pipeline
bert_asr = pipeline(
    "automatic-speech-recognition", model="alvanlii/wav2vec2-BERT-cantonese", device="cuda"
)
text = pipe(file)["text"]

Training Hyperparameters

learning_rate: 1e-4
train_batch_size: 4 (on 1 3090)
eval_batch_size: 1
gradient_accumulation_steps: 32
total_train_batch_size: 32x4=128
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_warmup_steps: 500

Training Results

Training Loss	Step	Validation Loss	CER
2.416	1200	1.615	0.4246
1.313	4200	0.9049	0.2745
1.090	7200	0.7463	0.2388
0.907	9600	0.6820	0.2172