Moshi

Speech-text foundation model for full-duplex real-time spoken dialogue with neural audio codec.

Open SourceSelf HostedOffline CapableGPU Required (24GB+ VRAM)
0.0 (0)

About

Moshi is a speech-text foundation model from Kyutai for full-duplex real-time spoken dialogue. It processes two simultaneous audio streams while predicting text tokens, and bundles Mimi, a streaming neural audio codec running at 12.5 Hz with 1.1 kbps. The repository ships PyTorch, MLX, and Rust backends covering research, Apple Silicon, and production deployments.

Reviews (0)

Leave a Review

No reviews yet. Be the first to review!

Details

Price
Free
Platform
Local/Desktop
Difficulty
Advanced (4/5)
License
Apache-2.0
Minimum VRAM
24 GB
Added
May 7, 2026

Related Tools

Featured

Lightweight and expressive TTS model with 82M parameters for fast local inference.

Open SourceSelf HostedOffline
Easy
4.0 (1)

Open-source TTS model by Resemble AI with emotion and accent control.

Open SourceSelf HostedOfflineGPU 4GB+
Easy
0.0 (0)

Expressive zero-shot TTS model by Resemble AI with emotion and accent control.

Open SourceSelf HostedOfflineGPU 4GB+
Intermediate
0.0 (0)

Singing voice conversion model based on VITS and SoftVC for voice-to-voice transfer.

Open SourceSelf HostedOfflineGPU 6GB+
Advanced
0.0 (0)

Zero-shot TTS model with high naturalness and speaker similarity.

Open SourceSelf HostedOfflineGPU 6GB+
Intermediate
0.0 (0)
Featured

Transformer-based text-to-audio model by Suno that generates speech, music, and sound effects.

Open SourceSelf HostedOfflineGPU 4GB+
Intermediate
0.0 (0)
Browse all Text-to-Speech (TTS) tools