Fine-tuned XLSR-53 large model for speech recognition in Italian
Fine-tuned facebook/wav2vec2-large-xlsr-53 on Italian using the train and validation splits of Common Voice 6.1. When using this model, make sure that your speech input is sampled at 16kHz.
This model has been fine-tuned thanks to the GPU credits generously given by the OVHcloud :)
The script used for training can be found here: https://github.com/jonatasgrosman/wav2vec2-sprint
Usage
The model can be used directly (without a language model) as follows...
Using the HuggingSound library:
from huggingsound import SpeechRecognitionModel
model = SpeechRecognitionModel("jonatasgrosman/wav2vec2-large-xlsr-53-italian")
audio_paths = ["/path/to/file.mp3", "/path/to/another_file.wav"]
transcriptions = model.transcribe(audio_paths)
Writing your own inference script:
import torch
import librosa
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
LANG_ID = "it"
MODEL_ID = "jonatasgrosman/wav2vec2-large-xlsr-53-italian"
SAMPLES = 10
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)
Reference | Prediction |
---|---|
POI LEI MORร. | POI LEI MORร |
IL LIBRO HA SUSCITATO MOLTE POLEMICHE A CAUSA DEI SUOI CONTENUTI. | IL LIBRO HA SUSCITATO MOLTE POLEMICHE A CAUSA DEI SUOI CONTENUTI |
"FIN DALL'INIZIO LA SEDE EPISCOPALE ร STATA IMMEDIATAMENTE SOGGETTA ALLA SANTA SEDE." | FIN DALL'INIZIO LA SEDE EPISCOPALE ร STATA IMMEDIATAMENTE SOGGETTA ALLA SANTA SEDE |
IL VUOTO ASSOLUTO? | IL VUOTO ASSOLUTO |
DOPO ALCUNI ANNI, EGLI DECISE DI TORNARE IN INDIA PER RACCOGLIERE ALTRI INSEGNAMENTI. | DOPO ALCUNI ANNI EGLI DECISE DI TORNARE IN INDIA PER RACCOGLIERE ALTRI INSEGNAMENTI |
SALVATION SUE | SALVATION SOO |
IN QUESTO MODO, DECIO OTTENNE IL POTERE IMPERIALE. | IN QUESTO MODO DECHO OTTENNE IL POTERE IMPERIALE |
SPARTA NOVARA ACQUISISCE IL TITOLO SPORTIVO PER GIOCARE IN PRIMA CATEGORIA. | PARCANOVARACFILISCE IL TITOLO SPORTIVO PER GIOCARE IN PRIMA CATEGORIA |
IN SEGUITO, KYGO E SHEAR HANNO PROPOSTO DI CONTINUARE A LAVORARE SULLA CANZONE. | IN SEGUITO KIGO E SHIAR HANNO PROPOSTO DI CONTINUARE A LAVORARE SULLA CANZONE |
ALAN CLARKE | ALAN CLARK |
Evaluation
- To evaluate on
mozilla-foundation/common_voice_6_0
with splittest
python eval.py --model_id jonatasgrosman/wav2vec2-large-xlsr-53-italian --dataset mozilla-foundation/common_voice_6_0 --config it --split test
- To evaluate on
speech-recognition-community-v2/dev_data
python eval.py --model_id jonatasgrosman/wav2vec2-large-xlsr-53-italian --dataset speech-recognition-community-v2/dev_data --config it --split validation --chunk_length_s 5.0 --stride_length_s 1.0
Citation
If you want to cite this model you can use this:
@misc{grosman2021xlsr53-large-italian,
title={Fine-tuned {XLSR}-53 large model for speech recognition in {I}talian},
author={Grosman, Jonatas},
howpublished={\url{https://huggingface.co/jonatasgrosman/wav2vec2-large-xlsr-53-italian}},
year={2021}
}
- Downloads last month
- 70,411
Datasets used to train jonatasgrosman/wav2vec2-large-xlsr-53-italian
Spaces using jonatasgrosman/wav2vec2-large-xlsr-53-italian 6
Evaluation results
- Test WER on Common Voice itself-reported9.410
- Test CER on Common Voice itself-reported2.290
- Test WER (+LM) on Common Voice itself-reported6.910
- Test CER (+LM) on Common Voice itself-reported1.830
- Dev WER on Robust Speech Event - Dev Dataself-reported21.780
- Dev CER on Robust Speech Event - Dev Dataself-reported7.940
- Dev WER (+LM) on Robust Speech Event - Dev Dataself-reported15.820
- Dev CER (+LM) on Robust Speech Event - Dev Dataself-reported6.830