语言: ml
数据集:

• Indic TTS 马拉雅拉姆语语音语料库
• Openslr 马拉雅拉姆语语音语料库
• SMC 马拉雅拉姆语语音语料库
• IIIT-H 印度语语音数据库
  评估指标:
• 词错误率 (wer)
  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: 马拉雅拉姆语 XLSR Wav2Vec2 大型 53
  结果:
  • 任务:
    名称: 语音识别
    类型: 自动语音识别
    数据集:
    名称: 使用上述所有数据集的组合测试集
    类型: 自定义
    参数: ml
    指标:
    • 名称: 测试词错误率
      类型: wer
      值: 28.43

Wav2Vec2-Large-XLSR-53-ml

基于 facebook/wav2vec2-large-xlsr-53 在马拉雅拉姆语 (ml) 上微调，使用了 Indic TTS 马拉雅拉姆语语音语料库 (通过 Kaggle)、Openslr 马拉雅拉姆语语音语料库、SMC 马拉雅拉姆语语音语料库 和 IIIT-H 印度语语音数据库。训练模型所用的笔记本可在此处找到。使用此模型时，请确保语音输入采样率为16kHz。

使用方法

该模型可直接使用（无需语言模型）如下：

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

test_dataset = <加载组合数据集的测试集> # 加载数据集的详细信息见评估部分  

processor = Wav2Vec2Processor.from_pretrained("gvs/wav2vec2-large-xlsr-malayalam")  
model = Wav2Vec2ForCTC.from_pretrained("gvs/wav2vec2-large-xlsr-malayalam")  

resampler = torchaudio.transforms.Resample(48_000, 16_000)  

# 预处理数据集  
# 需要将音频文件读取为数组  
def speech_file_to_array_fn(batch):  
  speech_array, sampling_rate = torchaudio.load(batch["path"])  
  batch["speech"] = resampler(speech_array).squeeze().numpy()  
  return batch  

test_dataset = test_dataset.map(speech_file_to_array_fn)  
inputs = processor(test_dataset["speech"][:2], 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)  

print("预测:", processor.batch_decode(predicted_ids))  
print("参考:", test_dataset["sentence"])  

评估

该模型可在组合自定义数据集的测试数据上进行如下评估。有关数据集准备的更多详情，请查看本文末尾提到的笔记本。

import torch  
import torchaudio  
from datasets import load_dataset, load_metric  
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor  
import re  
from datasets import load_dataset, load_metric  
from pathlib import Path  

# 需要使用本文末尾提到的笔记本创建自定义数据集  
data_dir = Path('<自定义数据集路径>')  

dataset_folders = {  
    'iiit': 'iiit_mal_abi',  
    'openslr': 'openslr',  
    'indic-tts': 'indic-tts-ml',  
    'msc-reviewed': 'msc-reviewed-speech-v1.0+20200825',  
}  

# 设置数据集目录  
openslr_male_dir = data_dir / dataset_folders['openslr'] / 'male'  
openslr_female_dir = data_dir / dataset_folders['openslr'] / 'female'  
iiit_dir = data_dir / dataset_folders['iiit']  
indic_tts_male_dir = data_dir / dataset_folders['indic-tts'] / 'male'  
indic_tts_female_dir = data_dir / dataset_folders['indic-tts'] / 'female'  
msc_reviewed_dir = data_dir / dataset_folders['msc-reviewed']  

# 加载数据集  
openslr_male = load_dataset("json", data_files=[f"{str(openslr_male_dir.absolute())}/sample_{i}.json" for i in range(2023)], split="train")  
openslr_female = load_dataset("json", data_files=[f"{str(openslr_female_dir.absolute())}/sample_{i}.json" for i in range(2103)], split="train")  
iiit = load_dataset("json", data_files=[f"{str(iiit_dir.absolute())}/sample_{i}.json" for i in range(1000)], split="train")  
indic_tts_male = load_dataset("json", data_files=[f"{str(indic_tts_male_dir.absolute())}/sample_{i}.json" for i in range(5649)], split="train")  
indic_tts_female = load_dataset("json", data_files=[f"{str(indic_tts_female_dir.absolute())}/sample_{i}.json" for i in range(2950)], split="train")  
msc_reviewed = load_dataset("json", data_files=[f"{str(msc_reviewed_dir.absolute())}/sample_{i}.json" for i in range(1541)], split="train")  

# 创建20%的测试集，并设置随机种子  
test_size = 0.2  
random_seed=1  
openslr_male_splits = openslr_male.train_test_split(test_size=test_size, seed=random_seed)  
openslr_female_splits = openslr_female.train_test_split(test_size=test_size, seed=random_seed)  
iiit_splits = iiit.train_test_split(test_size=test_size, seed=random_seed)  
indic_tts_male_splits = indic_tts_male.train_test_split(test_size=test_size, seed=random_seed)  
indic_tts_female_splits = indic_tts_female.train_test_split(test_size=test_size, seed=random_seed)  
msc_reviewed_splits = msc_reviewed.train_test_split(test_size=test_size, seed=random_seed)  

# 获取组合测试数据集  
split_list = [openslr_male_splits, openslr_female_splits, indic_tts_male_splits, indic_tts_female_splits, msc_reviewed_splits, iiit_splits]  
test_dataset = datasets.concatenate_datasets([split['test'] for split in split_list])  

wer = load_metric("wer")  

processor = Wav2Vec2Processor.from_pretrained("gvs/wav2vec2-large-xlsr-malayalam")  
model = Wav2Vec2ForCTC.from_pretrained("gvs/wav2vec2-large-xlsr-malayalam")  
model.to("cuda")  

resamplers = {  
    48000: torchaudio.transforms.Resample(48_000, 16_000),  
}  

chars_to_ignore_regex = '[\\\\,\\\\?\\\\.\\\\!\\\\-\\\\;\\\\:\\\\"\\\\“\\\\%\\\\‘\\\\”\\\\�Utrnle\\\\_]'  
unicode_ignore_regex = r'[\\\\u200e]'  

# 预处理数据集  
# 需要将音频文件读取为数组  
def speech_file_to_array_fn(batch):  
    batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"])  
    batch["sentence"] = re.sub(unicode_ignore_regex, '', batch["sentence"])  
    speech_array, sampling_rate = torchaudio.load(batch["path"])  
    # 如果采样率不是16kHz，则重新采样  
    if sampling_rate != 16000:  
        batch["speech"] = resamplers[sampling_rate](speech_array).squeeze().numpy()  
    else:  
        batch["speech"] = speech_array.squeeze().numpy()  
    # 如果存在多个维度，取第一个  
    if batch["speech"].ndim > 1:  
        batch["speech"] = batch["speech"][0]  
    return batch  

test_dataset = test_dataset.map(speech_file_to_array_fn)  

# 预处理数据集  
# 需要将音频文件读取为数组  
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("cuda"), attention_mask=inputs.attention_mask.to("cuda")).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("词错误率: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))  

测试结果 (词错误率): 28.43 %

训练

使用 Indic TTS 马拉雅拉姆语语音语料库 (通过 Kaggle)、Openslr 马拉雅拉姆语语音语料库、SMC 马拉雅拉姆语语音语料库 和 IIIT-H 印度语语音数据库 创建了一个组合数据集。数据集下载后，使用此笔记本转换为 HF 数据集格式。

训练和评估所用的笔记本可在此处找到。