Preeti uses Piper TTS, an open-source neural text-to-speech engine, to convert text to natural-sounding speech. We've made several improvements to the standard models to enhance Nepali language pronunciation and intonation.
The goal of this project is to provide Nepali audio generation, as well as a shareable content-generation tool that makes it easy to add voice-overs to videos.
This is an experimental project focused on Nepali language technology. The voice models have been tweaked to improve pronunciation of Devanagari characters and create more natural speech patterns. For more information, or feedbacks, please contact ishanshrestha@gmail.com
Will be continuously improving Preeti and would love to hear your thoughts. Please share your experience, suggestions, or report any issues you encounter.