Wav2vec2 Large Xlsr 53 Polish

🚀 Fine-tuned XLSR-53 large model for speech recognition in Polish

This project presents a fine - tuned facebook/wav2vec2-large-xlsr-53 model for Polish speech recognition. It is trained and validated on Common Voice 6.1. Ensure that your speech input is sampled at 16kHz when using this model. The model is fine - tuned with the GPU credits from OVHcloud. The training script can be found at: https://github.com/jonatasgrosman/wav2vec2-sprint.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

Using the HuggingSound library:

from huggingsound import SpeechRecognitionModel

model = SpeechRecognitionModel("jonatasgrosman/wav2vec2-large-xlsr-53-polish")
audio_paths = ["/path/to/file.mp3", "/path/to/another_file.wav"]

transcriptions = model.transcribe(audio_paths)

Advanced Usage

Writing your own inference script:

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

LANG_ID = "pl"
MODEL_ID = "jonatasgrosman/wav2vec2-large-xlsr-53-polish"
SAMPLES = 5

test_dataset = load_dataset("common_voice", LANG_ID, split=f"test[:{SAMPLES}]")

processor = Wav2Vec2Processor.from_pretrained(MODEL_ID)
model = Wav2Vec2ForCTC.from_pretrained(MODEL_ID)

# Preprocessing the datasets.
# We need to read the audio files as arrays
def speech_file_to_array_fn(batch):
    speech_array, sampling_rate = librosa.load(batch["path"], sr=16_000)
    batch["speech"] = speech_array
    batch["sentence"] = batch["sentence"].upper()
    return batch

test_dataset = test_dataset.map(speech_file_to_array_fn)
inputs = processor(test_dataset["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)
predicted_sentences = processor.batch_decode(predicted_ids)

for i, predicted_sentence in enumerate(predicted_sentences):
    print("-" * 100)
    print("Reference:", test_dataset[i]["sentence"])
    print("Prediction:", predicted_sentence)

Here is a comparison table of reference and prediction:

📚 Documentation

Evaluation

1. To evaluate on mozilla-foundation/common_voice_6_0 with split test

python eval.py --model_id jonatasgrosman/wav2vec2-large-xlsr-53-polish --dataset mozilla-foundation/common_voice_6_0 --config pl --split test

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

python eval.py --model_id jonatasgrosman/wav2vec2-large-xlsr-53-polish --dataset speech-recognition-community-v2/dev_data --config pl --split validation --chunk_length_s 5.0 --stride_length_s 1.0

Model Information

Results

• Automatic Speech Recognition on Common Voice pl:
  • Test WER: 14.21
  • Test CER: 3.49
  • Test WER (+LM): 10.98
  • Test CER (+LM): 2.93
• Automatic Speech Recognition on Robust Speech Event - Dev Data:
  • Dev WER: 33.18
  • Dev CER: 15.92
  • Dev WER (+LM): 29.31
  • Dev CER (+LM): 15.17

📄 License

This project is licensed under the Apache - 2.0 license.

📖 Citation

If you want to cite this model you can use this:

@misc{grosman2021xlsr53-large-polish,
  title={Fine-tuned {XLSR}-53 large model for speech recognition in {P}olish},
  author={Grosman, Jonatas},
  howpublished={\url{https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-polish}},
  year={2021}
}