Wav2vec2 Indonesian Javanese Sundanese

🚀 Multilingual Speech Recognition for Indonesian Languages

This project focuses on multilingual speech recognition for Indonesian languages. The model built here addresses the challenge of accurately transcribing speech in multiple Indonesian languages. It offers a practical solution for applications requiring speech - to - text conversion in these languages, enhancing accessibility and communication.

🚀 Quick Start

This is the model built for the project Multilingual Speech Recognition for Indonesian Languages. It is a fine - tuned facebook/wav2vec2-large-xlsr-53 model on the Indonesian Common Voice dataset, High - quality TTS data for Javanese - SLR41, and High - quality TTS data for Sundanese - SLR44 datasets.

We also provide a live demo to test the model.

When using this model, make sure that your speech input is sampled at 16kHz.

✨ Features

• Multilingual Support: Supports languages such as Indonesian (id), Javanese (jv), and Sundanese (sun).
• Fine - Tuned Model: Based on the pre - trained facebook/wav2vec2 - large - xlsr - 53 model, fine - tuned on specific Indonesian language datasets.
• Live Demo: Allows users to quickly test the model's performance.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

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

test_dataset = load_dataset("common_voice", "id", split="test[:2%]")

processor = Wav2Vec2Processor.from_pretrained("indonesian-nlp/wav2vec2-indonesian-javanese-sundanese")
model = Wav2Vec2ForCTC.from_pretrained("indonesian-nlp/wav2vec2-indonesian-javanese-sundanese")

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"])

Advanced Usage

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

test_dataset = load_dataset("common_voice", "id", split="test")
wer = load_metric("wer")

processor = Wav2Vec2Processor.from_pretrained("indonesian-nlp/wav2vec2-indonesian-javanese-sundanese")
model = Wav2Vec2ForCTC.from_pretrained("indonesian-nlp/wav2vec2-indonesian-javanese-sundanese") 
model.to("cuda")

chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\%\‘\'\”\�]'

resampler = torchaudio.transforms.Resample(48_000, 16_000)

# Preprocessing the datasets.
# We need to read the audio files as arrays
def speech_file_to_array_fn(batch):
    batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
    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)

# Preprocessing the datasets.
# We need to read the audio files as arrays
def evaluate(batch):
    inputs = processor(batch["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)
    batch["pred_strings"] = processor.batch_decode(pred_ids)
    return batch

result = test_dataset.map(evaluate, batched=True, batch_size=8)

print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))

Evaluation Results

Test Result: 11.57 %

📚 Documentation

Model Information

Model Index

• Name: Wav2Vec2 Indonesian Javanese and Sundanese by Indonesian NLP
• Results:
  • Task: Automatic Speech Recognition
    • Dataset: Common Voice 6.1
      • Test WER: 4.056
      • Test CER: 1.472
    • Dataset: Common Voice 7
      • Test WER: 4.492
      • Test CER: 1.577
    • Dataset: Robust Speech Event - Dev Data
      • Test WER: 48.94
    • Dataset: Robust Speech Event - Test Data
      • Test WER: 68.95

🔧 Technical Details

The Common Voice train, validation, and ... datasets were used for training as well as ... and ... # TODO

The script used for training can be found here (will be available soon)

📄 License

This project is licensed under the apache - 2.0 license.