import torch
import torchaudio

from transformers import AutoModelForCTC, Wav2Vec2Processor

processor = Wav2Vec2Processor.from_pretrained("bhuang/wav2vec2-xls-r-1b-voxpopuli-fr")
model = AutoModelForCTC.from_pretrained("bhuang/wav2vec2-xls-r-1b-voxpopuli-fr").cuda()

# path to your audio file
wav_path = "example.wav"
waveform, sample_rate = torchaudio.load(wav_path)
waveform = waveform.squeeze(axis=0)  # mono

# resample
if sample_rate != 16_000:
    resampler = torchaudio.transforms.Resample(sample_rate, 16_000)
    waveform = resampler(waveform)

# normalize
input_dict = processor(waveform, sampling_rate=16_000, return_tensors="pt")

with torch.inference_mode():
    logits = model(input_dict.input_values.to("cuda")).logits

# decode
predicted_ids = torch.argmax(logits, dim=-1)
predicted_sentence = processor.batch_decode(predicted_ids)[0]

To use on a local audio file with the language model

python

import torch
import torchaudio

from transformers import AutoModelForCTC, Wav2Vec2ProcessorWithLM

processor_with_lm = Wav2Vec2ProcessorWithLM.from_pretrained("bhuang/wav2vec2-xls-r-1b-voxpopuli-fr")
model = AutoModelForCTC.from_pretrained("bhuang/wav2vec2-xls-r-1b-voxpopuli-fr").cuda()

model_sampling_rate = processor_with_lm.feature_extractor.sampling_rate

# path to your audio file
wav_path = "example.wav"
waveform, sample_rate = torchaudio.load(wav_path)
waveform = waveform.squeeze(axis=0)  # mono

# resample
if sample_rate != 16_000:
    resampler = torchaudio.transforms.Resample(sample_rate, 16_000)
    waveform = resampler(waveform)

# normalize
input_dict = processor_with_lm(waveform, sampling_rate=16_000, return_tensors="pt")

with torch.inference_mode():
    logits = model(input_dict.input_values.to("cuda")).logits

predicted_sentence = processor_with_lm.batch_decode(logits.cpu().numpy()).text[0]

Evaluation

To evaluate on polinaeterna/voxpopuli

bash

python eval.py \
  --model_id "bhuang/wav2vec2-xls-r-1b-voxpopuli-fr" \
  --dataset "polinaeterna/voxpopuli" \
  --config "fr" \
  --split "test" \
  --log_outputs \
  --outdir "outputs/results_polinaeterna_voxpopuli_with_lm"

To evaluate on mozilla-foundation/common_voice_9_0

bash

python eval.py \
  --model_id "bhuang/wav2vec2-xls-r-1b-voxpopuli-fr" \
  --dataset "mozilla-foundation/common_voice_9_0" \
  --config "fr" \
  --split "test" \
  --log_outputs \
  --outdir "outputs/results_mozilla-foundatio_common_voice_9_0_with_lm"

To evaluate on speech-recognition-community-v2/dev_data

bash

python eval.py \
  --model_id "bhuang/wav2vec2-xls-r-1b-voxpopuli-fr" \
  --dataset "speech-recognition-community-v2/dev_data" \
  --config "fr" \
  --split "validation" \
  --chunk_length_s 5.0 \
  --stride_length_s 1.0 \
  --log_outputs \
  --outdir "outputs/results_speech-recognition-community-v2_dev_data_with_lm"

bofenghuang/wav2vec2-xls-r-1b-voxpopuli-fr

Fine-tuned Wav2Vec2 XLS-R 1B model for ASR in French

Usage

Evaluation

bofenghuang/
wav2vec2-xls-r-1b-voxpopuli-fr