automatic-speech-recognition mozilla-foundation/common_voice_8_0 robust-speech-event xlsr-fine-tuning-week hf-asr-leaderboard

wav2vec2-xls-r-300m-cs-cv8

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the common_voice 8.0 dataset. It achieves the following results on the evaluation set:

Usage

The model can be used directly (without a language model) as follows:

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

test_dataset = load_dataset("mozilla-foundation/common_voice_8_0", "sk", split="test[:2%]")

processor = Wav2Vec2Processor.from_pretrained("comodoro/wav2vec2-xls-r-300m-sk-cv8")
model = Wav2Vec2ForCTC.from_pretrained("comodoro/wav2vec2-xls-r-300m-sk-cv8")

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[:2]["speech"], 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[:2]["sentence"])

Evaluation

The model can be evaluated using the attached eval.py script:

python eval.py --model_id comodoro/wav2vec2-xls-r-300m-sk-cv8 --dataset mozilla-foundation/common_voice_8_0 --split test --config sk

Training and evaluation data

The Common Voice 8.0 train and validation datasets were used for training

Training hyperparameters

The following hyperparameters were used during training:

Framework versions