Overview / Usage

The project involve** creating a bot which will mimic your voice.** Generally when the voice assistant speaks we can easily understand that it is the voice of robot but in this software the bot will replicate your voice learning from your various voice recording. This bot will** learn from the waveform produced by your voice using your various voice recordings **and will pronounce the words written by you as your text.

This bot will solve the main problem for the Deaf people who cannot speak. This bot will produce the waveform using the superposition of the waveform produced by their parents learning from their voice recordings giving the person a new voice. So the bot will become voice for that person which he can pronounce by writing it in text. This can also be used in translation where your can pronounce other languages in your voice using the translator. This project will also help voice assistant like google assistant and amazon alexa to improve their voice and mimic their owner's voice. It can also be used by the employee to attend the phone of clients by speaking in the voice of manager in its absent.

This research is still in its learning phase and can be used in various of application if get successful. This would be great advantageous not only to people but also to the company due to its originality in voice.

​

Other links

• End-to-end Text-to-speech for Low-resource Languages
• MelNet: A Generative Model for Audio in the Frequency Domain
• Deep generative model of raw audio waveforms