注意

点击此处下载完整的示例代码

使用 Wav2Vec2 进行语音识别¶

作者：Moto Hira

本教程演示如何使用 Wav2Vec 2.0 的预训练模型 [论文] 进行语音识别。

概述¶

语音识别的过程如下所示。

从音频波形中提取声学特征
逐帧估计声学特征的类别
从类别概率序列中生成假设

Torchaudio 提供了对预训练权重及其相关信息（如期望的采样率和类别标签）的便捷访问。它们被打包在一起，可以在 torchaudio.pipelines 模块下找到。

准备¶

import torch
import torchaudio

print(torch.__version__)
print(torchaudio.__version__)

torch.random.manual_seed(0)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

print(device)

2.10.0.dev20251013+cu126
2.8.0a0+1d65bbe
cuda

import IPython
import matplotlib.pyplot as plt
from torchaudio.utils import _download_asset

SPEECH_FILE = _download_asset("tutorial-assets/Lab41-SRI-VOiCES-src-sp0307-ch127535-sg0042.wav")

100.0%

创建管道¶

首先，我们将创建一个执行特征提取和分类的 Wav2Vec2 模型。

Torchaudio 中有两种类型的 Wav2Vec2 预训练权重。一种是针对 ASR 任务进行微调的，另一种是没有进行微调的。

Wav2Vec2 (和 HuBERT) 模型以自监督方式训练。它们首先仅使用音频进行表示学习，然后使用额外的标签对特定任务进行微调。

未经微调的预训练权重也可以为其他下游任务进行微调，但本教程不涵盖此内容。

这里我们将使用 torchaudio.pipelines.WAV2VEC2_ASR_BASE_960H。

在 torchaudio.pipelines 中有多个预训练模型可用。有关它们如何训练的详细信息，请查阅文档。

bundle 对象提供了实例化模型和其他信息的接口。采样率和类别标签如下所示。

bundle = torchaudio.pipelines.WAV2VEC2_ASR_BASE_960H

print("Sample Rate:", bundle.sample_rate)

print("Labels:", bundle.get_labels())

Sample Rate: 16000
Labels: ('-', '|', 'E', 'T', 'A', 'O', 'N', 'I', 'H', 'S', 'R', 'D', 'L', 'U', 'M', 'W', 'C', 'F', 'G', 'Y', 'P', 'B', 'V', 'K', "'", 'X', 'J', 'Q', 'Z')

模型可以按如下方式构建。此过程将自动获取预训练权重并将其加载到模型中。

model = bundle.get_model().to(device)

print(model.__class__)

Downloading: "https://download.pytorch.org/torchaudio/models/wav2vec2_fairseq_base_ls960_asr_ls960.pth" to /root/.cache/torch/hub/checkpoints/wav2vec2_fairseq_base_ls960_asr_ls960.pth

0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
0%
1%
1%
1%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
9%
0%
0%
1%
1%
1%
2%
2%
2%
3%
3%
3%
4%
4%
4%
5%
5%
5%
6%
6%
6%
7%
7%
7%
8%
8%
8%
9%
9%
0%
0%
0%
<class 'torchaudio.models.wav2vec2.model.Wav2Vec2Model'>

加载数据¶

我们将使用来自 VOiCES 数据集的语音数据，该数据集已获得 Creative Commons BY 4.0 许可。

IPython.display.Audio(SPEECH_FILE)

要加载数据，我们使用 torchaudio.load()。

如果采样率与管道期望的不同，则可以使用 torchaudio.functional.resample() 进行重采样。

注意

torchaudio.functional.resample() 也适用于 CUDA 张量。
当对同一组采样率进行多次重采样时，使用 torchaudio.transforms.Resample 可能会提高性能。

waveform, sample_rate = torchaudio.load(SPEECH_FILE)
waveform = waveform.to(device)

if sample_rate != bundle.sample_rate:
    waveform = torchaudio.functional.resample(waveform, sample_rate, bundle.sample_rate)

提取声学特征¶

下一步是从音频中提取声学特征。

注意

针对 ASR 任务进行微调的 Wav2Vec2 模型可以一步完成特征提取和分类，但为了教程的目的，我们也在此展示了如何执行特征提取。

with torch.inference_mode():
    features, _ = model.extract_features(waveform)

返回的特征是张量列表。每个张量都是一个 Transformer 层的输出。

fig, ax = plt.subplots(len(features), 1, figsize=(16, 4.3 * len(features)))
for i, feats in enumerate(features):
    ax[i].imshow(feats[0].cpu(), interpolation="nearest")
    ax[i].set_title(f"Feature from transformer layer {i+1}")
    ax[i].set_xlabel("Feature dimension")
    ax[i].set_ylabel("Frame (time-axis)")
fig.tight_layout()

Feature from transformer layer 1, Feature from transformer layer 2, Feature from transformer layer 3, Feature from transformer layer 4, Feature from transformer layer 5, Feature from transformer layer 6, Feature from transformer layer 7, Feature from transformer layer 8, Feature from transformer layer 9, Feature from transformer layer 10, Feature from transformer layer 11, Feature from transformer layer 12

特征分类¶

一旦提取了声学特征，下一步就是将它们分类为一组类别。

Wav2Vec2 模型提供了一种一步完成特征提取和分类的方法。

with torch.inference_mode():
    emission, _ = model(waveform)

输出是 logits 的形式。它不是概率的形式。

让我们可视化一下。

plt.imshow(emission[0].cpu().T, interpolation="nearest")
plt.title("Classification result")
plt.xlabel("Frame (time-axis)")
plt.ylabel("Class")
plt.tight_layout()
print("Class labels:", bundle.get_labels())

Class labels: ('-', '|', 'E', 'T', 'A', 'O', 'N', 'I', 'H', 'S', 'R', 'D', 'L', 'U', 'M', 'W', 'C', 'F', 'G', 'Y', 'P', 'B', 'V', 'K', "'", 'X', 'J', 'Q', 'Z')

我们可以看到，在时间线上对某些标签存在强烈的指示。

生成转录¶

从标签概率序列中，我们现在想要生成转录。生成假设的过程通常称为“解码”。

解码比简单分类更复杂，因为特定时间步的解码可能会受到周围观测值的影响。

例如，考虑像 night 和 knight 这样的词。即使它们的先验概率分布不同（在典型的对话中，night 出现的频率远高于 knight），要准确生成包含 knight 的转录，例如 a knight with a sword，解码过程必须等到看到足够的上下文才能做出最终决定。

已经提出了许多解码技术，它们需要外部资源，例如单词词典和语言模型。

在本教程中，为简单起见，我们将执行贪婪解码，它不依赖于此类外部组件，并且在每个时间步简单地选择最佳假设。因此，不使用上下文信息，只能生成一个转录。

我们首先定义贪婪解码算法。

class GreedyCTCDecoder(torch.nn.Module):
    def __init__(self, labels, blank=0):
        super().__init__()
        self.labels = labels
        self.blank = blank

    def forward(self, emission: torch.Tensor) -> str:
        """Given a sequence emission over labels, get the best path string
        Args:
          emission (Tensor): Logit tensors. Shape `[num_seq, num_label]`.

        Returns:
          str: The resulting transcript
        """
        indices = torch.argmax(emission, dim=-1)  # [num_seq,]
        indices = torch.unique_consecutive(indices, dim=-1)
        indices = [i for i in indices if i != self.blank]
        return "".join([self.labels[i] for i in indices])

现在创建解码器对象并解码转录。

decoder = GreedyCTCDecoder(labels=bundle.get_labels())
transcript = decoder(emission[0])

让我们检查结果并再次收听音频。

print(transcript)
IPython.display.Audio(SPEECH_FILE)

I|HAD|THAT|CURIOSITY|BESIDE|ME|AT|THIS|MOMENT|

ASR 模型使用称为连接时序分类 (CTC) 的损失函数进行微调。CTC 损失的详细信息解释在此。在 CTC 中，空白标记 (ϵ) 是一个特殊标记，表示前一个符号的重复。在解码中，这些标记被简单地忽略。

结论¶

在本教程中，我们研究了如何使用 Wav2Vec2ASRBundle 进行声学特征提取和语音识别。构建模型和获取发射（emission）只需两行代码。

model = torchaudio.pipelines.WAV2VEC2_ASR_BASE_960H.get_model()
emission = model(waveforms, ...)

脚本总运行时间： ( 0 分钟 5.806 秒)

由 Sphinx-Gallery 生成的画廊