RealtimeTTS Releases

The 0.5.0 release optimizes performance by refactoring module initialization to use on-demand loading and lazy loading strategies.

Mar 28, 2025

This update enhances the OpenAI engine configuration options and introduces new voices, alongside improvements to error tracking and thread management.

Mar 24, 2025

A new OrpheusEngine is added to support the Orpheus-3B model, featuring emotive speech tags and low-latency streaming capabilities.

Mar 22, 2025

This release fixes a critical bug where asterisks in text would cause synthesis failures in the Kokoro engine.

Mar 19, 2025

This update adds a callback mechanism to track word-level timing during playback, supported by Azure and Kokoro engines.

Mar 17, 2025

Adds voice mixing capabilities to the KokoroEngine, allowing users to blend custom voices using a specific formula syntax.

Mar 9, 2025

Introduces a speed control parameter to the KokoroEngine, allowing users to adjust the playback rate of the TTS output.

Mar 9, 2025

Updates the CoquiEngine to support multiple voice files by allowing the `set_voice` method to accept a list of WAV references.

Feb 15, 2025

Fixes a critical bug where paused audio streams could not be successfully stopped.

Feb 9, 2025

Expands the Kokoro voice library to include Japanese and Chinese languages through a specific optional installation flag.

Feb 8, 2025

Updates the Kokoro engine dependency to version 0.7.3 to resolve specific issues identified in GitHub issue #259.

Feb 4, 2025

Simplifies KokoroEngine installation to be managed entirely by pip and upgrades to support Kokoro V1.0 models with a wider range of voices.

Feb 3, 2025

Introduces the new KokoroEngine to the library, enabling users to utilize the Kokoro 82M text-to-speech model with various voice options.

Jan 11, 2025

Adds granular control over audio buffering parameters to help mitigate stuttering issues, particularly on Unix-based systems.

Jan 6, 2025

Introduces the PiperEngine to support the Piper TTS model, requiring a separate installation of the Piper executable alongside the library.

Jan 2, 2025

This release updates core dependencies and introduces a seed parameter to the StyleTTS engine while addressing noise generation issues with short text.

Dec 14, 2024

Users can now utilize 48 kHz audio output with the Azure TTS engine and enjoy dynamic voice switching capabilities within the StyleTTS engine.

Dec 10, 2024

The library now supports the StyleTTS2 engine and has been updated to the latest version of Coqui-TTS.

Dec 7, 2024

This release introduces new language versions and translations for the 'Hello 算法' PDF book, expanding support for Ruby and optimizing existing code and translations.

Dec 6, 2024

This update focuses on performance optimization and bug fixes, specifically improving the testing utilities for the Edge TTS engine.

Nov 30, 2024