语言: 奥里亚语
数据集:

• OpenSLR
  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: Shyam Sunder Kumar的XLSR Wav2Vec2奥里亚语模型
  结果:
  • 任务:
    名称: 语音识别
    类型: 自动语音识别
    数据集:
    名称: OpenSLR
    类型: OpenSLR
    参数: or
    指标:
    • 名称: 测试词错误率
      类型: wer
      值: 68.75

Wav2Vec2-Large-XLSR-53-奥里亚语

基于facebook/wav2vec2-large-xlsr-53微调的奥里亚语模型，使用了低资源印度语言的多语言和语码转换ASR挑战赛数据。
使用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），示例如下：

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

test_dataset = load_dataset("common_voice", "or", split="test[:2%]")  
processor = Wav2Vec2Processor.from_pretrained("theainerd/wav2vec2-large-xlsr-53-odia")  
model = Wav2Vec2ForCTC.from_pretrained("theainerd/wav2vec2-large-xlsr-53-odia")  
resampler = torchaudio.transforms.Resample(48_000, 16_000)  

# 数据预处理  
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("预测结果:", processor.batch_decode(predicted_ids))  
print("参考文本:", test_dataset["sentence"][:2])  

评估

可在Common Voice的奥里亚语测试数据上按如下方式评估模型：

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

test_dataset = load_dataset("common_voice", "or", split="test")  
wer = load_metric("wer")  

processor = Wav2Vec2Processor.from_pretrained("theainerd/wav2vec2-large-xlsr-53-odia")  
model = Wav2Vec2ForCTC.from_pretrained("theainerd/wav2vec2-large-xlsr-53-odia")  
model.to("cuda")  

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

def speech_file_to_array_fn(batch):  
  batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()  
  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)  

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: 68.75%

训练

训练脚本参见奥里亚语ASR微调Wav2Vec2