语言: eo
数据集:

• common_voice
  指标:
• wer
  标签:
• audio
• automatic-speech-recognition
• speech
• xlsr-fine-tuning-week
  许可证: apache-2.0
  模型索引:
• 名称: XLSR Wav2Vec2 Esperanto by Charles Pierse 结果:
  • 任务: 名称: 语音识别 类型: automatic-speech-recognition 数据集: 名称: Common Voice eo 类型: common_voice 参数: eo 指标:
    • 名称: 测试WER 类型: wer 值: 12.31

Wav2Vec2-Large-XLSR-53-eo

基于facebook/wav2vec2-large-xlsr-53模型，使用Common Voice数据集对世界语（Esperanto）进行了微调。

使用此模型时，请确保语音输入采样率为16kHz。

使用方法

该模型可以直接使用（无需语言模型），如下所示：

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

test_dataset = load_dataset("common_voice", "eo", split="test[:2%]")
processor = Wav2Vec2Processor.from_pretrained("cpierse/wav2vec2-large-xlsr-53-esperanto")
model = Wav2Vec2ForCTC.from_pretrained("cpierse/wav2vec2-large-xlsr-53-esperanto")

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

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

评估

可以在Common Voice的世界语测试数据上评估该模型，如下所示：

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

def chunked_wer(targets, predictions, chunk_size=None):
    if chunk_size is None: return jiwer.wer(targets, predictions)
    start = 0
    end = chunk_size
    H, S, D, I = 0, 0, 0, 0
    while start < len(targets):
        chunk_metrics = jiwer.compute_measures(targets[start:end], predictions[start:end])
        H = H + chunk_metrics["hits"]
        S = S + chunk_metrics["substitutions"]
        D = D + chunk_metrics["deletions"]
        I = I + chunk_metrics["insertions"]
        start += chunk_size
        end += chunk_size
    return float(S + D + I) / float(H + S + D)

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

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

chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\%\‘\”\�\„\«\(\»\)\’\']'
resampler = torchaudio.transforms.Resample(48_000, 16_000)

# 数据预处理
# 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"]).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)

# 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("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("WER: {:2f}".format(100 * chunked_wer(predictions=result["pred_strings"], targets=result["sentence"],chunk_size=2000)))

测试结果: 12.31%

训练

使用了Common Voice的train和validation数据集进行训练。