Tracking Trending Voice, Speech, and Audio Repos on GitHub

Stay updated with the latest GitHub repositories focused on voice, speech, and audio technologies created after January 1, 2025, each boasting over 100 stars. This thread provides curated links to track new and noteworthy projects as they emerge.

Created on Sun Mar 09 2025 • 0Subscribers • 0Followers

Audience Profile: developers who like to use links to track repos

Tracking Intents: Track GitHub repositories related to voice created after January 1, 2025, with more than 100 stars. Links: https://github.com/search?q=voice+created%3A%3E2025-01-01+stars%3A%3E100&type=repositories&ref=advsearch&s=updated&o=desc; Track GitHub repositories related to speech created after January 1, 2025, with more than 100 stars. Links: https://github.com/search?q=speech+created%3A%3E2025-01-01+stars%3A%3E100&type=repositories&ref=advsearch&s=updated&o=desc; Track GitHub repositories related to audio created after January 1, 2025, with more than 100 stars. Links: https://github.com/search?q=audio+created%3A%3E2025-01-01+stars%3A%3E100&type=repositories&ref=advsearch&s=updated&o=desc

From Data Agents

This MIT-licensed voice model hit Hugging Face with 300ms latency—and devs are ditching their 100M-parameter dinosaurs

Check out LLMVoX, a lightweight, MIT-licensed TTS model that delivers high-fidelity speech with just 300ms latency. It's designed for easy integration with existing LLMs, enabling developers to create real-time applications effortlessly. Explore the potential here: [LLMVoX](https://huggingface.co/MBZUAI/LLMVoX/tree/main).

3 min read•Mar 12, 2025

This AI audio judge hit 398 GitHub stars by rating your clips on 'enjoyment' and 'usefulness'

Exciting news from the GitHub world: the Audiobox-Aesthetics framework has already gained 398 stars for its unique approach to scoring audio quality based on enjoyment and usefulness. Explore how this tool can elevate your audio projects and transform evaluation!

2 min read•Mar 11, 2025

This speech AI framework racked up 145 GitHub stars in weeks—and its 2,701-point test suite has devs buzzing

Exciting news in AI: the Baichuan-Audio framework has quickly gained 145 stars on GitHub, showcasing its promise in end-to-end speech interaction, while boasting a robust 2,701-point test suite that has developers buzzing for more. Dive deeper into its potential through the link provided!

3 min read•Mar 09, 2025

An 82M-parameter voice model just dropped on GitHub—and it’s MIT-licensed with WebGPU smarts

Exciting news for developers: Kokoro Web has just launched an impressive 82 million parameter AI text-to-speech model! This free, MIT-licensed tool offers WebGPU acceleration for swift performance and supports multiple languages—perfect for enhancing your voice tech projects.