Wav2vec2 Base 10k Voxpopuli Ft Hr

🚀 Wav2Vec2-Base-VoxPopuli-Finetuned

本项目基于 Facebook的Wav2Vec2 基础模型，该模型在 VoxPopuli语料库 的10K无标签子集上进行预训练，并在克罗地亚语（hr）的转录数据上进行微调（更多信息请参考论文中的表1）。

🚀 快速开始

模型信息

相关论文

论文：VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation 作者：来自 Facebook AI 的 Changhan Wang、Morgane Riviere、Ann Lee、Anne Wu、Chaitanya Talnikar、Daniel Haziza、Mary Williamson、Juan Pino、Emmanuel Dupoux

更多信息请访问官方网站 here

💻 使用示例

基础用法

以下展示了如何在 Common Voice数据集 的样本上使用该模型进行推理：

#!/usr/bin/env python3
from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
from datasets import load_dataset
import torchaudio
import torch

# resample audio

# load model & processor
model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-10k-voxpopuli-ft-hr")
processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-base-10k-voxpopuli-ft-hr")

# load dataset
ds = load_dataset("common_voice", "hr", split="validation[:1%]")

# common voice does not match target sampling rate
common_voice_sample_rate = 48000
target_sample_rate = 16000

resampler = torchaudio.transforms.Resample(common_voice_sample_rate, target_sample_rate)


# define mapping fn to read in sound file and resample
def map_to_array(batch):
    speech, _ = torchaudio.load(batch["path"])
    speech = resampler(speech)
    batch["speech"] = speech[0]
    return batch


# load all audio files
ds = ds.map(map_to_array)

# run inference on the first 5 data samples
inputs = processor(ds[:5]["speech"], sampling_rate=target_sample_rate, return_tensors="pt", padding=True)

# inference
logits = model(**inputs).logits
predicted_ids = torch.argmax(logits, axis=-1)

print(processor.batch_decode(predicted_ids))

📄 许可证

本项目采用CC BY-NC 4.0许可证。