MossVoice Studio
Expressive zero-shot voice cloning and high-fidelity speech generation powered by OpenMOSS MOSS-TTS Transformer nodes inside ComfyUI.
1
Reference Voice Sample
Provide 5 to 15 seconds of clean, single-speaker audio
Drop audio file here or browse
Supports WAV, MP3, FLAC, OGG, M4A, or WEBM (5–20s clean speech)
00:00
Click the microphone to record your voice directly from browser.
⚡
LANfactory Voice Preset
Studio reference audio: "Hello from LAN factory! The audio synthesis system is online."
Voice Library Presets:
💡 MOSS Zero-Shot Advantage: MOSS-TTS uses dual acoustic codebooks to capture vocal prosody, delivery, and timbre directly from audio — no transcript required!
2
What Should MOSS Say?
Type your script or choose an inspiration take below
Inspiration Scripts:
🌿 MOSS Expression Engine: Emotional prosody is driven by audio temperature (2.0+) and punctuation (`!`, `?`, `...`). For maximum emotional acting without reference voice anchoring, synthesize in Pure Narration mode (click ✕ on reference audio).
🎵 Ambient Sound & Sound Event Tags (Optional) ▼
3
MOSS-TTS Synthesis Parameters
Configure model weights, language, and sampling dynamics
🚀 Running on GPU CUDA
Audio Temperature
1.70
Recommended: 1.70 for v1.5
Audio Top-P
0.80
Nucleus codebook cutoff: 0.80
Voice Output Studio
Synthesized neural speech playback and audio export
Ready for Synthesis
No MOSS Speech Synthesized Yet
Load a voice reference or write a prompt, configure your acting style, and hit "Synthesize MOSS Speech" to start the ComfyUI pipeline.
Model
MOSS-TTS v1.5
Duration
--
Speed
--
Recent Session Takes
Generated audio takes from this session will appear here for comparison.