In development · Media & Audio AI
Dhwani
An AI audio studio: invent sounds, generate speech and build voices.
This project is in development. Features described are planned and being built; it is not yet a live product, and details may change before release. Screens and demonstrations use sample data. Read our disclaimer.
Overview
Dhwani (Sanskrit for "sound") is an AI platform for discovering and making audio. It creates original sounds mathematically, evaluates them, and saves the good ones as exact recipes that recreate the same sound every time. You can describe a sound in plain words and the AI suggests the parameters. Speech, dialogue and voice studios sit alongside, and it all runs privately on your own machine or local network.
Key features
- Sound Lab: create original sounds and save them as recipes you can reproduce
- "Describe a sound": a local AI turns your words into sound settings
- Speech Studio and Dialogue: generated speech and conversations with several speakers
- Voices and Recording Studio: record from phones or laptops over Wi-Fi
- Library of samples and saved sounds
- Waveform preview and playback
- User accounts with admin and member roles
- Coming soon Music, songs and singing
Roadmap
- Coming soon Music composition and MIDI
- Coming soon Audio generated from a script, scene by scene
Want a system like this for your business?
Tell us how your business works. We'll suggest what to build first and how to grow it.