wayne-xyz / Voice-Translator

Speech Translate app and server that can generate audio based on your own voice

Voice Translator App and Server

Overview

The Voice Translator App and Server is a project that enables users to interactively translate spoken language using their own voice. It provides a seamless communication experience by allowing users to engage in real-time conversations with people who speak different languages.

Features

• Voice Recognition: This App can recognize user's voice and turn to the text content.
• Language Translation: This App can send user's speak content to server to processing translation, At present, only Chinese and English are supported.
• Real-time Interaction: It has a relatively high response speed and supports real-time face-to-face conversations for two individuals.
• Voice mimic: Generated translation is baed on user own voice. It utilize the ML on Server.

Framwork & tools

iOS

Swift, Custom UIView, Speech framework,Grand Central Dispatch(GCD), AVFoundation.

Server

Python, Fastapi, Pytorch, Cuda, XTTS, argostranslate.

Showcase

• Custom animation when you say something ,the circle will be changed by your voice power

• Custom animation when data is loading the ring will turn circle

![Video]

Usage

1. Support Python 3.11.5
2. Support iOS 15.6
3. Additionally need XTTS-v2 folder