rany2/edge-tts
Edge Tts
edge-tts is a command line interface and text-to-speech engine that converts written text into audio files using the Microsoft Edge online synthesis service. It functions as a client for generating high-quality speech and managing the conversion of text to audio.
The project provides utilities for generating synchronized SRT subtitle files by tracking word and sentence boundaries during synthesis. It also includes a voice profile discovery system to browse a catalog of available synthetic voices based on gender and personality traits.
Users can customize vocal characteristics by adjusting the pitch, rate, and volume of the output. The system supports real-time audio streaming for immediate playback without requiring local file storage, allowing for the simultaneous streaming of synthesized speech and synchronized text captions.
Features
- Text-to-Speech - Converts written text into high-quality audio files using the Microsoft Edge online synthesis service.
- Speech Synthesis Gateways - Offloads text-to-speech conversion to a remote cloud engine via a synthesis gateway.
- Speech Synthesis & TTS - Provides a command line interface for performing text-to-speech synthesis via Microsoft Edge.
- Websocket Connection Managers - Maintains persistent duplex WebSocket connections for sending text and receiving binary audio streams.
- Voice Profile Managers - Provides a system for querying catalogs and selecting specific vocal identity profiles.
- Vocal Characteristic Adjustments - Allows customization of pitch, rate, and volume to create specific vocal characteristics.
- Voice Discovery Interfaces - Provides a discovery interface to browse a catalog of synthetic voices based on gender and traits.
- Speech Parameter Configuration - Allows adjustment of voice profiles, playback rates, volume, and pitch.
- Vocal Tone Customization - Provides tools to modify speaker characteristics and tone to change synthesized audio quality.
- Timestamped Subtitle Generators - Generates synchronized SRT files by parsing timing metadata for word and sentence boundaries.
- Audio Playback Engines - Streams synthesized audio for immediate playback without requiring the audio to be saved to disk.
- Synthesis Playback Clients - Functions as a client for playing synthesized speech in real time without local file storage.
- Instant Streaming Playback - Streams synthesized audio directly to a media player for instant listening without disk storage.
- Real-Time Audio Streaming Buffers - Processes raw audio data packets in real time via streaming buffers for immediate playback.
- Speech Processing - Library for using Microsoft Edge's TTS service.
- Speech Synthesis - Wrapper for Microsoft Edge's text-to-speech service.
