🚀 Wav2Vec2-Large-XLSR-53-Cantonese
This model is fine-tuned from facebook/wav2vec2-large-xlsr-53 on Cantonese using the Common Voice. It's designed for automatic speech recognition tasks.
Model Information
Property |
Details |
Model Type |
wav2vec2-large-xlsr-cantonese |
Training Data |
Common Voice (train, validation sets) |
Metrics |
CER (Test CER: 15.36) |
License |
apache-2.0 |
Important Note
⚠️ Important Note
When using this model, make sure that your speech input is sampled at 16kHz.
🚀 Quick Start
This fine-tuned model is based on facebook/wav2vec2-large-xlsr-53 and trained on Cantonese with the Common Voice.
💻 Usage Examples
Basic Usage
import torch
import torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
test_dataset = load_dataset("common_voice", "zh-HK", split="test[:2%]")
processor = Wav2Vec2Processor.from_pretrained("ctl/wav2vec2-large-xlsr-cantonese")
model = Wav2Vec2ForCTC.from_pretrained("ctl/wav2vec2-large-xlsr-cantonese")
resampler = torchaudio.transforms.Resample(48_000, 16_000)
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])
📚 Documentation
Evaluation
The model can be evaluated on the Chinese (Hong Kong) test data of Common Voice as follows:
!pip install jiwer
import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import re
import argparse
lang_id = "zh-HK"
model_id = "ctl/wav2vec2-large-xlsr-cantonese"
chars_to_ignore_regex = '[\,\?\.\!\-\;\:"\“\%\‘\”\�\.\⋯\!\-\:\–\。\》\,\)\,\?\;\~\~\…\︰\,\(\」\‧\《\﹔\、\—\/\,\「\﹖\·\']'
test_dataset = load_dataset("common_voice", f"{lang_id}", split="test")
cer = load_metric("cer")
processor = Wav2Vec2Processor.from_pretrained(f"{model_id}")
model = Wav2Vec2ForCTC.from_pretrained(f"{model_id}")
model.to("cuda")
resampler = torchaudio.transforms.Resample(48_000, 16_000)
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)
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=16)
print("CER: {:2f}".format(100 * cer.compute(predictions=result["pred_strings"], references=result["sentence"])))
Test Result: 15.51 %
Training
The Common Voice train
and validation
sets were used for training. The script used for training will be posted here.
📄 License
This model is released under the apache-2.0 license.