Wav2vec2 Large Xlsr 53 Dutch

🚀 wav2vec2-large-xlsr-53-Dutch

該模型基於 facebook/wav2vec2-large-xlsr-53，使用 Common Voice 數據集針對荷蘭語進行了微調。使用此模型時，請確保語音輸入的採樣率為 16kHz。

✨ 主要特性

• 多領域標籤支持：涵蓋音頻、自動語音識別、語音等多個領域標籤。
• 特定語言微調：專門針對荷蘭語進行微調，適用於荷蘭語語音識別任務。
• 模型評估結果明確：在 Common Voice 荷蘭語測試數據上有明確的字錯誤率（WER）評估結果。

📦 安裝指南

文檔未提及具體安裝步驟，可參考 Hugging Face 相關模型的通用安裝方式。

💻 使用示例

基礎用法

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

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

高級用法

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

test_dataset = load_dataset("common_voice", "nl", split="test")
wer = load_metric("wer")
processor = Wav2Vec2Processor.from_pretrained("MehdiHosseiniMoghadam/wav2vec2-large-xlsr-53-Dutch")
model = Wav2Vec2ForCTC.from_pretrained("MehdiHosseiniMoghadam/wav2vec2-large-xlsr-53-Dutch")
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(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"])))

模型評估結果

測試結果：26.494162 %

📚 詳細文檔

訓練信息

訓練使用了 Common Voice 的 train 和 validation 數據集。

📄 許可證

本模型使用 Apache-2.0 許可證。

模型信息表格