Vaani

🚀 Speechbrain Audio Classification Model

This model is designed for audio classification, supporting multiple languages including English, Hindi, Tamil, Telugu, and Marathi.

🚀 Quick Start

To get started with the model, you can use the following code:

from speechbrain.inference.classifiers import EncoderClassifier
import time

# Initialize the classifier
classifier = EncoderClassifier.from_hparams(source="panchajanya-ai/vaani")

# Use CUDA if available
classifier = EncoderClassifier.from_hparams(source="panchajanya-ai/vaani", run_opts={"device": "cuda"})

# Start the timer
start_time = time.time()

# Classify an audio file
out_prob, score, index, text_lab = classifier.classify_file('sample_audio.wav')

# End the timer
end_time = time.time()

# Print the results
print("Probabilities:", out_prob)
print("Score:", score)
print("Index:", index)
print("Label:", text_lab)
print("Time taken:", end_time - start_time, "seconds")

📄 License

This project is licensed under the Apache-2.0 license.

📋 Model Information