语言: 阿拉伯语
数据集:

• 通用语音库
• 阿拉伯语语音语料库
  评估指标:
• 词错误率(WER)
• 字符错误率(CER)
  标签:
• 音频
• 自动语音识别
• 语音
• XLSR微调周
  许可证: Apache-2.0
  模型索引:
• 名称: Jonatas Grosman的XLSR Wav2Vec2阿拉伯语模型
  结果:
  • 任务:
    名称: 语音识别
    类型: 自动语音识别
    数据集:
    名称: 通用语音库阿拉伯语版
    类型: 通用语音库
    参数: 阿拉伯语
    评估指标:
    • 名称: 测试WER
      类型: WER
      值: 39.59
    • 名称: 测试CER
      类型: CER
      值: 18.18

针对阿拉伯语语音识别微调的XLSR-53大模型

基于facebook/wav2vec2-large-xlsr-53模型，使用通用语音库6.1和阿拉伯语语音语料库的训练集和验证集进行微调。
使用该模型时，请确保语音输入采样率为16kHz。

本模型的微调得益于OVHcloud慷慨提供的GPU算力支持 :)

训练脚本详见: https://github.com/jonatasgrosman/wav2vec2-sprint

使用方式

该模型可直接使用（无需语言模型）如下...

使用HuggingSound库:

from huggingsound import SpeechRecognitionModel  

model = SpeechRecognitionModel("jonatasgrosman/wav2vec2-large-xlsr-53-arabic")  
audio_paths = ["/path/to/file.mp3", "/path/to/another_file.wav"]  

transcriptions = model.transcribe(audio_paths)  

自定义推理脚本:

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

LANG_ID = "ar"  
MODEL_ID = "jonatasgrosman/wav2vec2-large-xlsr-53-arabic"  
SAMPLES = 10  

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

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

# 数据预处理  
# 需将音频文件读取为数组  
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("参考文本:", test_dataset[i]["sentence"])  
    print("预测结果:", predicted_sentence)  

评估

可在通用语音库阿拉伯语测试集上按如下方式评估模型表现。

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

LANG_ID = "ar"  
MODEL_ID = "jonatasgrosman/wav2vec2-large-xlsr-53-arabic"  
DEVICE = "cuda"  

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

test_dataset = load_dataset("common_voice", LANG_ID, split="test")  

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)  

# 数据预处理  
# 需将音频文件读取为数组  
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)  

# 评估函数  
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)  

predictions = [x.upper() for x in result["pred_strings"]]  
references = [x.upper() for x in result["sentence"]]  

print(f"WER: {wer.compute(predictions=predictions, references=references, chunk_size=1000) * 100}")  
print(f"CER: {cer.compute(predictions=predictions, references=references, chunk_size=1000) * 100}")  

测试结果:

下表展示了模型的词错误率(WER)和字符错误率(CER)。我们于2021年5月14日使用相同评估脚本对比了其他模型。注意由于评估脚本差异，下表结果可能与已报告数据存在出入。

引用

若需引用本模型，请使用以下格式:

@misc{grosman2021xlsr53-large-arabic,  
  title={针对阿拉伯语语音识别微调的{XLSR}-53大模型},  
  author={Grosman, Jonatas},  
  howpublished={\url{https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-arabic}},  
  year={2021}  
}