Wav2vec2 Large Xlsr 53 Swedish

🚀 Wav2Vec2-Large-XLSR-53-瑞典語模型

本項目是對 facebook/wav2vec2-large-xlsr-53 模型進行瑞典語微調後的成果，使用了 NST Swedish Dictation 數據集。使用該模型時，請確保語音輸入的採樣率為 16kHz。

注意：為獲得最佳性能，建議使用我們更新的模型 wav2vec2-large-voxrex-swedish。

🚀 快速開始

此模型可直接使用（無需語言模型），以下是使用示例：

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

test_dataset = load_dataset("common_voice", "sv-SE", split="test[:2%]")

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

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])

✨ 主要特性

• 基於 facebook/wav2vec2-large-xlsr-53 模型進行瑞典語微調。
• 可直接用於瑞典語的自動語音識別任務。

💻 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", "sv-SE", split="test[:2%]")

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

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])

📚 詳細文檔

評估

該模型可在 Common Voice 的瑞典語測試數據上進行評估，示例代碼如下：

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

test_dataset = load_dataset("common_voice", "sv-SE", split="test")
wer = load_metric("wer")

processor = Wav2Vec2Processor.from_pretrained("KBLab/wav2vec2-large-xlsr-53-swedish")
model = Wav2Vec2ForCTC.from_pretrained("KBLab/wav2vec2-large-xlsr-53-swedish")
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"])))
print("CER: {:2f}".format(100 * wer.compute(predictions=[" ".join(list(entry)) for entry in result["pred_strings"]], references=[" ".join(list(entry)) for entry in result["sentence"]])))

字錯率（WER）：14.298610% 字符錯誤率（CER）：4.925294%

訓練

模型的訓練過程如下：

1. 首先，使用包含來自各個廣播電臺的 1000 小時瑞典語語音語料，對 XLSR 模型進行 50 個週期的預訓練。
2. 其次，使用 NST Swedish Dictation 和 Common Voice 數據集進行微調。
3. 最後，僅使用 Common Voice 數據集進行最終微調。訓練使用了 Fairseq 腳本。

📄 許可證

本項目採用 Apache 2.0 許可證。

📦 數據集和指標

📦 模型信息