Wav2vec2 Large Xlsr As

🚀 瓦2向量2-大型-XLSR-53-阿萨姆语模型

本项目基于Common Voice数据集，在阿萨姆语上对facebook/wav2vec2-large-xlsr-53模型进行了微调。该模型可用于阿萨姆语的自动语音识别任务，为语音识别领域在阿萨姆语方面提供了有效的解决方案。

🚀 快速开始

在使用此模型时，请确保您的语音输入采样率为 16kHz。

✨ 主要特性

• 模型类型：微调后的瓦2向量2（Wav2Vec2）大型XLSR - 53模型。
• 训练数据：Common Voice阿萨姆语的训练集和验证集。
• 评估指标：词错误率（WER），测试集上的WER为 69.63%。

💻 使用示例

基础用法

import torch
import torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
test_dataset = load_dataset("common_voice", "as", split="test[:2%]")
processor = Wav2Vec2Processor.from_pretrained("anuragshas/wav2vec2-large-xlsr-as")
model = Wav2Vec2ForCTC.from_pretrained("anuragshas/wav2vec2-large-xlsr-as")
resampler = torchaudio.transforms.Resample(48_000, 16_000)
# Preprocessing the datasets.
# We need to read the aduio files as arrays
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("Prediction:", processor.batch_decode(predicted_ids))
print("Reference:", test_dataset["sentence"][:2])

高级用法

import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import re
test_dataset = load_dataset("common_voice", "as", split="test")
wer = load_metric("wer")
processor = Wav2Vec2Processor.from_pretrained("anuragshas/wav2vec2-large-xlsr-as")
model = Wav2Vec2ForCTC.from_pretrained("anuragshas/wav2vec2-large-xlsr-as")
model.to("cuda")
chars_to_ignore_regex = '[\\,\\?\\.\\!\\-\\;\\:\\\"\\“\\%\\”\\়\\।]'
resampler = torchaudio.transforms.Resample(48_000, 16_000)
# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
    batch["sentence"] = re.sub('’ ',' ',batch["sentence"])
    batch["sentence"] = re.sub(' ‘',' ',batch["sentence"])
    batch["sentence"] = re.sub('’|‘','\'',batch["sentence"])
    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)
# Preprocessing the datasets.
# We need to read the aduio files as arrays
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 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))

测试结果：69.63 %

🔧 技术细节

本模型在训练时使用了Common Voice的train和validation数据集进行训练，以提升模型在阿萨姆语语音识别任务上的性能。

📄 许可证

本项目采用 Apache - 2.0 许可证。