README.md

---
language:
- ja
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_11_0
metrics:
- wer
- cer
model-index:
- name: uniTKU-hubert-japanese-asr
  results:
  - task:
      type: automatic-speech-recognition
      name: Speech Recognition
    dataset:
      name: common_voice_11_0
      type: common_voice
      args: ja
    metrics:
    - type: wer
      value: 27.511982
      name: Test WER
    - type: cer
      value: 11.563649
      name: Test CER
---

# uniTKU-hubert-japanese-asr

This model was fine-tuned on a dataset provided by uniTKU, and it has maintained the original performance metrics on the [common_voice_11_0 dataset](https://huggingface.co/datasets/mozilla-foundation/common_voice_11_0/viewer/ja).

This model can only predict Hiragana.

## Training Procedure

Fine-tuning on the uniTKU dataset led to the following results:

| Step  | Training Loss | Validation Loss | WER    |
|-------|---------------|-----------------|--------|
| 100  | 1.127100      | 1.089644        | 0.668508|
| 200  | 0.873500      | 0.682353        | 0.508287|
| 300  | 0.786200      | 0.482965        | 0.397790|
| 400  | 0.670400      | 0.345377        | 0.381215|
| 500  | 0.719500      | 0.387554        | 0.337017|
| 600  | 0.707700      | 0.371083        | 0.292818|
| 700  | 0.658300      | 0.236447        | 0.243094|
| 800  | 0.611100      | 0.207679        | 0.193370|

### Training hyperparameters

The training hyperparameters remained consistent throughout the fine-tuning process:

- learning_rate: 1e-4
- train_batch_size: 16
- eval_batch_size: 16
- gradient_accumulation_steps: 2
- max_steps: 800
- lr_scheduler_type: linear

### How to evaluate the model

```python
from transformers import HubertForCTC, Wav2Vec2Processor
from datasets import load_dataset
import torch
import torchaudio
import librosa
import numpy as np
import re
import MeCab
import pykakasi
from evaluate import load

model = HubertForCTC.from_pretrained('TKU410410103/uniTKU-hubert-japanese-asr')
processor = Wav2Vec2Processor.from_pretrained("TKU410410103/uniTKU-hubert-japanese-asr")

# load dataset
test_dataset = load_dataset('mozilla-foundation/common_voice_11_0', 'ja', split='test')
remove_columns = [col for col in test_dataset.column_names if col not in ['audio', 'sentence']]
test_dataset = test_dataset.remove_columns(remove_columns)

# resample
def process_waveforms(batch):
    speech_arrays = []
    sampling_rates = []

    for audio_path in batch['audio']:
        speech_array, _ = torchaudio.load(audio_path['path'])
        speech_array_resampled = librosa.resample(np.asarray(speech_array[0].numpy()), orig_sr=48000, target_sr=16000)
        speech_arrays.append(speech_array_resampled)
        sampling_rates.append(16000)

    batch["array"] = speech_arrays
    batch["sampling_rate"] = sampling_rates

    return batch

# hiragana
CHARS_TO_IGNORE = [",", "?", "¿", ".", "!", "¡", ";", "；", ":", '""', "%", '"', "�", "ʿ", "·", "჻", "~", "՞",
          "؟", "،", "।", "॥", "«", "»", "„", "“", "”", "「", "」", "‘", "’", "《", "》", "(", ")", "[", "]",
          "{", "}", "=", "`", "_", "+", "<", ">", "…", "–", "°", "´", "ʾ", "‹", "›", "©", "®", "—", "→", "。",
          "、", "﹂", "﹁", "‧", "～", "﹏", "，", "｛", "｝", "（", "）", "［", "］", "【", "】", "‥", "〽",
          "『", "』", "〝", "〟", "⟨", "⟩", "〜", "：", "！", "？", "♪", "؛", "/", "\\", "º", "−", "^", "'", "ʻ", "ˆ"]
chars_to_ignore_regex = f"[{re.escape(''.join(CHARS_TO_IGNORE))}]"

wakati = MeCab.Tagger("-Owakati")
kakasi = pykakasi.kakasi()
kakasi.setMode("J","H")
kakasi.setMode("K","H")
kakasi.setMode("r","Hepburn")
conv = kakasi.getConverter()

def prepare_char(batch):
    batch["sentence"] = conv.do(wakati.parse(batch["sentence"]).strip())
    batch["sentence"] = re.sub(chars_to_ignore_regex,'', batch["sentence"]).strip()
    return batch


resampled_eval_dataset = test_dataset.map(process_waveforms, batched=True, batch_size=50, num_proc=4)
eval_dataset = resampled_eval_dataset.map(prepare_char, num_proc=4)

# begin the evaluation process
wer = load("wer")
cer = load("cer")

def evaluate(batch):
    inputs = processor(batch["array"], sampling_rate=16_000, return_tensors="pt", padding=True)
    with torch.no_grad():
        logits = model(inputs.input_values.to(device), attention_mask=inputs.attention_mask.to(device)).logits
    pred_ids = torch.argmax(logits, dim=-1)
    batch["pred_strings"] = processor.batch_decode(pred_ids)
    return batch

columns_to_remove = [column for column in eval_dataset.column_names if column != "sentence"]
batch_size = 16
result = eval_dataset.map(evaluate, remove_columns=columns_to_remove, batched=True, batch_size=batch_size)

wer_result = wer.compute(predictions=result["pred_strings"], references=result["sentence"])
cer_result = cer.compute(predictions=result["pred_strings"], references=result["sentence"])

print("WER: {:2f}%".format(100 * wer_result))
print("CER: {:2f}%".format(100 * cer_result))
```

### Test results
The final model was evaluated as follows:

On uniTKU Dataset:
- WER: 19.003370%
- CER: 11.027523%

On common_voice_11_0:
- WER: 27.511982%
- CER: 11.563649%
  
### Framework versions

- Transformers 4.39.1
- Pytorch 2.2.1+cu118
- Datasets 2.17.1