LLM Reference

MOSS-TTS Models by MOSI AI

MOSI AIApache 2.0Open sourceOpen SourceAudio
1 model2026

Last refreshed 2026-06-29. Next refresh: weekly.

Details

ResearcherMOSI AI
LicenseApache 2.0OSI-approved
Commercial useCommercial use: permitted
Models1
Released2026

About

MOSS-TTS is an open-source speech and sound generation model family from MOSI AI and the OpenMOSS team, designed for high-fidelity, high-expressiveness text-to-speech across complex real-world scenarios. The family covers stable long-form speech, multi-speaker dialogue, voice and character design, environmental sound effects, and real-time streaming TTS. Models include MOSS-TTS v1.0, MOSS-TTS-v1.5, MOSS-TTS-Nano, MOSS-SoundEffect, MOSS-TTSD, and MOSS-VoiceGenerator.

Decision facts

Best fit
audiotext to speech
Capability starting point
MOSS-TTS-v1.5
Lowest tracked input
Not tracked
Closest related family
K2 Horizon

Current Variants

Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.

1 in view

Use when the workload needs text to speech, 8B parameters, and audio.

2026-05text to speech8B parametersaudio

Release Timeline

1 release group
2026-05
1 current
MOSS-TTS-v1.5
text to speech8B parametersaudio
Current

Specifications(1 models)

MOSS-TTS model specifications comparison
ModelReleasedParameters
MOSS-TTS-v1.52026-058B