Skip to main content

Ctrl+K

M* v0.2.1 Documentation

Getting Started

Installation
Quickstart

User Guide

Serving
Metrics reference
Using a Server
Evaluation and Benchmarks

Reference

Architecture
Supported Models
Speech recognition: Whisper and Qwen3-ASR
API Reference
Environment variables

Contributing

Adding a New Model

Repository
Show source
Open issue

.rst

mstar.model.voxcpm2.components

mstar.model.voxcpm2.components#

Modules

AudioVAE V2 of VoxCPM2: 16 kHz waveform -> 25 Hz latents -> 48 kHz waveform.

previous

mstar.model.voxcpm2

next

mstar.model.voxcpm2.components.audio_vae

By M* Team

© Copyright 2025-2026, M* Team.

Last updated on Nov 03, 2026.