Wav2vec2 Large Xlsr 53 Irish

🚀 Wav2Vec2-Large-XLSR-53-Irish

本項目基於 Common Voice 數據集，對愛爾蘭語的 facebook/wav2vec2-large-xlsr-53 模型進行了微調。使用該模型時，請確保語音輸入的採樣率為 16kHz。

🚀 快速開始

本模型是在愛爾蘭語數據上對 facebook/wav2vec2-large-xlsr-53 進行微調得到的。使用此模型時，請確保語音輸入的採樣率為 16kHz。

✨ 主要特性

• 微調模型：基於 facebook/wav2vec2-large-xlsr-53 在愛爾蘭語的 Common Voice 數據集上進行微調。
• 採樣率要求：語音輸入需採樣為 16kHz。

📦 安裝指南

文檔未提及安裝步驟，故跳過此章節。

💻 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", "ga-IE", split="test[:2%]") #TODO: replace {lang_id} in your language code here. Make sure the code is one of the *ISO codes* of [this](https://huggingface.co/languages) site.

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

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", "ga-IE", split="test") 
wer = load_metric("wer")

processor = Wav2Vec2Processor.from_pretrained("cpierse/wav2vec2-large-xlsr-53-irish") 
model = Wav2Vec2ForCTC.from_pretrained("cpierse/wav2vec2-large-xlsr-53-irish")
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"])))

📚 詳細文檔

模型信息

測試結果

測試結果：43.06 %

📄 許可證

本項目採用 Apache-2.0 許可證。