README.md

metadata

language: ja
datasets:
  - common_voice
metrics:
  - wer
  - cer
tags:
  - audio
  - automatic-speech-recognition
  - speech
  - xlsr-fine-tuning-week
license: apache-2.0
model-index:
  - name: XLSR Wav2Vec2 Japanese by Jonatas Grosman
    results:
      - task:
          name: Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Common Voice ja
          type: common_voice
          args: ja
        metrics:
          - name: Test WER
            type: wer
            value: 93.35
          - name: Test CER
            type: cer
            value: 29.24

Wav2Vec2-Large-XLSR-53-Japanese

Fine-tuned facebook/wav2vec2-large-xlsr-53 on Japanese using the Common Voice and CSS10. When using this model, make sure that your speech input is sampled at 16kHz.

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

Usage

The model can be used directly (without a language model) as follows:

import torch
import librosa
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor

LANG_ID = "ja"
MODEL_ID = "jonatasgrosman/wav2vec2-large-xlsr-53-japanese"
SAMPLES = 5

test_dataset = load_dataset("common_voice", LANG_ID, split=f"test[:{SAMPLES}]")

processor = Wav2Vec2Processor.from_pretrained(MODEL_ID)
model = Wav2Vec2ForCTC.from_pretrained(MODEL_ID)

# Preprocessing the datasets.
# We need to read the audio files as arrays
def speech_file_to_array_fn(batch):
    speech_array, sampling_rate = librosa.load(batch["path"], sr=16_000)
    batch["speech"] = speech_array
    batch["sentence"] = batch["sentence"].upper()
    return batch

test_dataset = test_dataset.map(speech_file_to_array_fn)
inputs = processor(test_dataset["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)

with torch.no_grad():
    logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits

predicted_ids = torch.argmax(logits, dim=-1)
predicted_sentences = processor.batch_decode(predicted_ids)

for i, predicted_sentence in enumerate(predicted_sentences):
    print("-" * 100)
    print("Reference:", test_dataset[i]["sentence"])
    print("Prediction:", predicted_sentence)

Reference	Prediction
祖母は、おおむね機嫌よく、サイコロをころがしている。	都ぼは重い記念よくさいこところがしている
財布をなくしたので、交番へ行きます。	財布王なクしたので、交番へへ行きます す
飲み屋のおやじ、旅館の主人、医者をはじめ、交際のある人にきいてまわったら、みんな、私より収入が多いはずなのに、税金は安い。	ノみアのやじ、旅館の筋時に、医者を初め、交際なる人に聞いて廻ったら、みんな、私しより周入が多い弾ず脱に、制金は安すい
新しい靴をはいて出かけます。	新しに靴をはいてかけます
このためプラズマ中のイオンや電子の持つ平均運動エネルギーを温度で表現することがある	このため、プラズマ中の医本や、電手のもつ平均運動をエネルギーを穏で、表現することがある

Evaluation

The model can be evaluated as follows on the Japanese test data of Common Voice.

import torch
import re
import librosa
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor

LANG_ID = "ja"
MODEL_ID = "jonatasgrosman/wav2vec2-large-xlsr-53-japanese"
DEVICE = "cuda"
MAX_SAMPLES = 8000

CHARS_TO_IGNORE = [",", "?", "¿", ".", "!", "¡", ";", ":", '""', "%", '"', "�", "ʿ", "·", "჻", "~", "՞", 
                   "؟", "،", "।", "॥", "«", "»", "„", "“", "”", "「", "」", "‘", "’", "《", "》", "(", ")", "[", "]",
                   "=", "`", "_", "+", "<", ">", "…", "–", "°", "´", "ʾ", "‹", "›", "©", "®", "—", "→", "。"]

test_dataset = load_dataset("common_voice", LANG_ID, split="test")
if len(test_dataset) > MAX_SAMPLES:
    test_dataset = test_dataset.select(range(MAX_SAMPLES))

wer = load_metric("wer.py") # https://github.com/jonatasgrosman/wav2vec2-sprint/blob/main/wer.py
cer = load_metric("cer.py") # https://github.com/jonatasgrosman/wav2vec2-sprint/blob/main/cer.py

chars_to_ignore_regex = f"[{re.escape(''.join(CHARS_TO_IGNORE))}]"

processor = Wav2Vec2Processor.from_pretrained(MODEL_ID)
model = Wav2Vec2ForCTC.from_pretrained(MODEL_ID)
model.to(DEVICE)

# Preprocessing the datasets.
# We need to read the audio files as arrays
def speech_file_to_array_fn(batch):
    with warnings.catch_warnings():
        warnings.simplefilter("ignore")
        speech_array, sampling_rate = librosa.load(batch["path"], sr=16_000)
    batch["speech"] = speech_array
    batch["sentence"] = re.sub(chars_to_ignore_regex, "", batch["sentence"]).upper()
    return batch

test_dataset = test_dataset.map(speech_file_to_array_fn)

# Preprocessing the datasets.
# We need to read the audio files as arrays
def evaluate(batch):
    inputs = processor(batch["speech"], 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

result = test_dataset.map(evaluate, batched=True, batch_size=8)

print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"], chunk_size=1000)))
print("CER: {:2f}".format(100 * cer.compute(predictions=result["pred_strings"], references=result["sentence"], chunk_size=1000)))

Test Result:

• WER: 93.35%
• CER: 29.24%