---
title: wav2vec2-large-xlsr-53-dutch
canonical_url: "https://www.modelscope.cn/models/facebook/wav2vec2-large-xlsr-53-dutch"
md_url: "https://www.modelscope.cn/models/facebook/wav2vec2-large-xlsr-53-dutch.md"
repository: facebook/wav2vec2-large-xlsr-53-dutch
last_updated: 2025-05-20
license: apache-2.0
model_type:
  - wav2vec2
architectures:
  - Wav2Vec2ForCTC
library_name:
  - pytorch
  - transformer
frameworks:
  - pytorch
language:
  - nl
downloads: 331
stars: 0
tags:
  - speech
  - audio
  - automatic-speech-recognition
---

# wav2vec2-large-xlsr-53-dutch

> wav2vec2-large-xlsr-53-dutch - facebook 在 ModelScope 开源的模型。Evaluation on Common Voice NL Test

facebook/wav2vec2-large-xlsr-53-dutch 是 ModelScope 魔搭社区上的机器学习模型，采用 apache-2.0 许可。

- **Repository**: facebook/wav2vec2-large-xlsr-53-dutch
- **License**: apache-2.0
- **Tags**: speech, audio, automatic-speech-recognition
- **Downloads**: 331
- **Stars**: 0
- **Last updated**: 2025-05-20

Source: https://www.modelscope.cn/models/facebook/wav2vec2-large-xlsr-53-dutch

---

## Evaluation on Common Voice NL Test

```python
import torchaudio
from datasets import load_dataset, load_metric
from transformers import (
    Wav2Vec2ForCTC,
    Wav2Vec2Processor,
)
import torch
import re
import sys

model_name = "facebook/wav2vec2-large-xlsr-53-dutch"
device = "cuda"
chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"]'  # noqa: W605

model = Wav2Vec2ForCTC.from_pretrained(model_name).to(device)
processor = Wav2Vec2Processor.from_pretrained(model_name)

ds = load_dataset("common_voice", "nl", split="test", data_dir="./cv-corpus-6.1-2020-12-11")

resampler = torchaudio.transforms.Resample(orig_freq=48_000, new_freq=16_000)

def map_to_array(batch):
    speech, _ = torchaudio.load(batch["path"])
    batch["speech"] = resampler.forward(speech.squeeze(0)).numpy()
    batch["sampling_rate"] = resampler.new_freq
    batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower().replace("’", "'")
    return batch

ds = ds.map(map_to_array)


def map_to_pred(batch):
    features = processor(batch["speech"], sampling_rate=batch["sampling_rate"][0], padding=True, return_tensors="pt")
    input_values = features.input_values.to(device)
    attention_mask = features.attention_mask.to(device)
    with torch.no_grad():
        logits = model(input_values, attention_mask=attention_mask).logits
    pred_ids = torch.argmax(logits, dim=-1)
    batch["predicted"] = processor.batch_decode(pred_ids)
    batch["target"] = batch["sentence"]
    return batch

result = ds.map(map_to_pred, batched=True, batch_size=16, remove_columns=list(ds.features.keys()))

wer = load_metric("wer")
print(wer.compute(predictions=result["predicted"], references=result["target"]))
```

**Result**: 21.1 %