openmoss/moss-tts-nano

A 100M-parameter multilingual TTS model for real-time CPU inference, voice cloning, and 48 kHz stereo generation

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 1 hour ago
Added to GitGenius on September 15th, 2026
Created on April 10th, 2026
Open Issues & Pull Requests: 24 (+0)
GitHub issues: Enabled
Number of forks: 555
Total Stargazers: 4,384 (+1)
Total Subscribers: 25 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 17.9 hours
Mean response time: 5.8 days
90th percentile: 21.0 days
Tracked items: 74

Most active contributors

Sign in to see contributor activity.

How this project is maintained

About 9% of issues opened in the past year have never received a reply. 79% of issues opened in the past year have been closed, leaving a working backlog. Three people close 92% of everything that gets resolved.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 17
New in 7 days: 0
Closed in 7 days: 0
Avg open age: 86 days
Stale 30+ days: 16
Stale 90+ days: 6

Recent activity

Opened in 7 days: 0
Closed in 7 days: 0
Comments in 7 days: 0
Events in 7 days: 0

Top labels

  • help wanted (1)
  • question (1)

Most active issues this week

Detailed Description

MOSS-TTS-Nano is a text-to-speech model that generates speech from text in real time on CPU hardware with support for voice cloning and high-quality stereo output.

The tool addresses the need for efficient speech synthesis that runs locally without GPU acceleration. It achieves this through a 100-million-parameter architecture designed for multilingual support, enabling users to synthesize speech in multiple languages. The model supports voice cloning, allowing it to adapt to specific speaker characteristics. Output is generated at 48 kHz in stereo format, providing higher audio quality than typical TTS systems.

The project suits developers building applications that require on-device speech synthesis without cloud dependencies or GPU resources. It is particularly valuable for systems with CPU-only constraints or where latency and privacy are concerns. The multilingual capability makes it applicable to international applications. The voice cloning feature enables personalized speech generation without retraining the base model.

Development activity shows consistent engagement with the codebase. The project maintains active issue tracking and responds to user reports. Pull requests are reviewed and merged regularly, indicating ongoing refinement of the model and implementation. Documentation updates occur alongside feature development, suggesting attention to usability as the tool evolves.