Banglaasr

🚀 孟加拉語自動語音識別模型（Bangla ASR）

本項目是一個基於孟加拉語 Mozilla Common Voice 數據集訓練的自動語音識別（ASR）模型。它通過對 Whisper 模型進行微調，在約 400 小時的孟加拉語語音數據上進行訓練，實現了較低的單詞錯誤率，為孟加拉語語音處理提供了有效的解決方案。

🚀 快速開始

以下是使用該模型進行語音識別的基本步驟：

import os
import librosa
import torch
import torchaudio
import numpy as np

from transformers import WhisperTokenizer
from transformers import WhisperProcessor
from transformers import WhisperFeatureExtractor
from transformers import WhisperForConditionalGeneration

device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

mp3_path = "https://huggingface.co/bangla-speech-processing/BanglaASR/resolve/main/mp3/common_voice_bn_31515636.mp3"

model_path = "bangla-speech-processing/BanglaASR"

feature_extractor = WhisperFeatureExtractor.from_pretrained(model_path)
tokenizer = WhisperTokenizer.from_pretrained(model_path)
processor = WhisperProcessor.from_pretrained(model_path)
model = WhisperForConditionalGeneration.from_pretrained(model_path).to(device)

speech_array, sampling_rate = torchaudio.load(mp3_path, format="mp3")
speech_array = speech_array[0].numpy()
speech_array = librosa.resample(np.asarray(speech_array), orig_sr=sampling_rate, target_sr=16000)
input_features = feature_extractor(speech_array, sampling_rate=16000, return_tensors="pt").input_features

# batch = processor.feature_extractor.pad(input_features, return_tensors="pt")
predicted_ids = model.generate(inputs=input_features.to(device))[0]

transcription = processor.decode(predicted_ids, skip_special_tokens=True)

print(transcription)

💻 使用示例

基礎用法

上述代碼展示瞭如何使用該模型對單個音頻文件進行語音識別。只需指定音頻文件的路徑和模型的路徑，即可完成語音識別並輸出轉錄結果。

高級用法

在實際應用中，你可能需要處理多個音頻文件或進行批量處理。以下是一個簡單的示例，展示如何處理多個音頻文件：

import os
import librosa
import torch
import torchaudio
import numpy as np

from transformers import WhisperTokenizer
from transformers import WhisperProcessor
from transformers import WhisperFeatureExtractor
from transformers import WhisperForConditionalGeneration

device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

model_path = "bangla-speech-processing/BanglaASR"

feature_extractor = WhisperFeatureExtractor.from_pretrained(model_path)
tokenizer = WhisperTokenizer.from_pretrained(model_path)
processor = WhisperProcessor.from_pretrained(model_path)
model = WhisperForConditionalGeneration.from_pretrained(model_path).to(device)

# 音頻文件列表
audio_files = [
    "https://huggingface.co/bangla-speech-processing/BanglaASR/resolve/main/mp3/common_voice_bn_31515636.mp3",
    "https://huggingface.co/bangla-speech-processing/BanglaASR/resolve/main/mp3/common_voice_bn_31549899.mp3",
    "https://huggingface.co/bangla-speech-processing/BanglaASR/resolve/main/mp3/common_voice_bn_31617644.mp3"
]

for mp3_path in audio_files:
    speech_array, sampling_rate = torchaudio.load(mp3_path, format="mp3")
    speech_array = speech_array[0].numpy()
    speech_array = librosa.resample(np.asarray(speech_array), orig_sr=sampling_rate, target_sr=16000)
    input_features = feature_extractor(speech_array, sampling_rate=16000, return_tensors="pt").input_features

    predicted_ids = model.generate(inputs=input_features.to(device))[0]
    transcription = processor.decode(predicted_ids, skip_special_tokens=True)

    print(f"音頻文件: {mp3_path}")
    print(f"轉錄結果: {transcription}")
    print("-" * 50)

📚 詳細文檔

數據集

本模型使用了 Mozilla Common Voice 數據集，該數據集包含約 400 小時的音頻數據，其中訓練集有 40k 個樣本，驗證集有 7k 個樣本。所有樣本均為 MP3 格式。 如需瞭解更多關於數據集的信息，請點擊此處。

訓練模型信息

評估

模型的單詞錯誤率（Word Error Rate）為 4.58%。 如需瞭解更多評估信息，請查看GitHub 倉庫。

引用

@misc{BanglaASR ,
  title={Transformer Based Whisper Bangla ASR Model},
  author={Md Saiful Islam},
  howpublished={},
  year={2023}
}

📄 許可證

本項目採用 MIT 許可證。