Talha-2003 / urdu-tts

An open source urdu/hindi text-to-speech system with voice cloning capabilities

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Generative Urdu Speech Synthesis

License: MIT Inference in Colab

Announcement

The weights are opensourced here: https://huggingface.co/zohann/urdu-tts

Features

  • South-Asian Accent Voice Cloning: tortoise-tts better captures european and american accents but when a non native speaker voice is cloned it americanizes that voice and completely fails. This model has been finetuned on a south-asian accent dataset, so it performs very well in cloning voices with a south-asian accent. to better capture and reproduce Indian accents, enabling accurate voice cloning for a variety of Indian English accents. It offers improved voice quality and natural-sounding speech synthesis for a more authentic experience.
  • Urdu Text-to-Speech: The model also includes support for Urdu text-to-speech, it can understand the arabic-urdu script and produce speech in urdu based on the text.

Results

Results and audios samples here: https://ahmedhanzala.github.io/urdu-tts/

Reference

License

This project is licensed under the MIT License. Feel free to use and modify the code according to your needs.

About

An open source urdu/hindi text-to-speech system with voice cloning capabilities

License:MIT License


Languages

Language:Jupyter Notebook 100.0%