đ Wav2Vec2-Large-XLSR-53-Latvian
This model is fine - tuned from facebook/wav2vec2-large-xlsr-53 on Latvian using the Common Voice dataset. It's designed for automatic speech recognition tasks.
đ Key Information
Property |
Details |
Model Type |
Wav2Vec2-Large-XLSR-53-Latvian |
Datasets |
common_voice |
Metrics |
wer |
Tags |
audio, automatic - speech - recognition, speech, xlsr - fine - tuning - week |
License |
apache - 2.0 |
đ Model Index
- Name: Latvian XLSR Wav2Vec2 Large 53 by Anton Lozhkov
- Results:
- Task:
- Name: Speech Recognition
- Type: automatic - speech - recognition
- Dataset:
- Name: Common Voice lv
- Type: common_voice
- Args: lv
- Metrics:
- Name: Test WER
- Type: wer
- Value: 26.89
đ Quick Start
When using this model, make sure that your speech input is sampled at 16kHz.
đģ Usage Examples
Basic Usage
import torch
import torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
test_dataset = load_dataset("common_voice", "lv", split="test[:2%]")
processor = Wav2Vec2Processor.from_pretrained("anton-l/wav2vec2-large-xlsr-53-latvian")
model = Wav2Vec2ForCTC.from_pretrained("anton-l/wav2vec2-large-xlsr-53-latvian")
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])
đ Documentation
Evaluation
The model can be evaluated as follows on the Latvian test data of Common Voice.
import torch
import torchaudio
import urllib.request
import tarfile
import pandas as pd
from tqdm.auto import tqdm
from datasets import load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
data_url = "https://voice-prod-bundler-ee1969a6ce8178826482b88e843c335139bd3fb4.s3.amazonaws.com/cv-corpus-6.1-2020-12-11/lv.tar.gz"
filestream = urllib.request.urlopen(data_url)
data_file = tarfile.open(fileobj=filestream, mode="r|gz")
data_file.extractall()
wer = load_metric("wer")
processor = Wav2Vec2Processor.from_pretrained("anton-l/wav2vec2-large-xlsr-53-latvian")
model = Wav2Vec2ForCTC.from_pretrained("anton-l/wav2vec2-large-xlsr-53-latvian")
model.to("cuda")
cv_test = pd.read_csv("cv-corpus-6.1-2020-12-11/lv/test.tsv", sep='\t')
clips_path = "cv-corpus-6.1-2020-12-11/lv/clips/"
def clean_sentence(sent):
sent = sent.lower()
sent = "".join(ch if ch.isalpha() else " " for ch in sent)
sent = " ".join(sent.split())
return sent
targets = []
preds = []
for i, row in tqdm(cv_test.iterrows(), total=cv_test.shape[0]):
row["sentence"] = clean_sentence(row["sentence"])
speech_array, sampling_rate = torchaudio.load(clips_path + row["path"])
resampler = torchaudio.transforms.Resample(sampling_rate, 16_000)
row["speech"] = resampler(speech_array).squeeze().numpy()
inputs = processor(row["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)
targets.append(row["sentence"])
preds.append(processor.batch_decode(pred_ids)[0])
print("WER: {:2f}".format(100 * wer.compute(predictions=preds, references=targets)))
Test Result: 26.89 %
Training
The Common Voice train
and validation
datasets were used for training.
đ License
This project is licensed under the apache - 2.0 license.