MyVoice-TTS - Feature Guide

Before You Start

Important information for new users

⚠

First-Time Startup

The first time you generate speech, MyVoice will automatically download the Qwen3-TTS models. This is a one-time download of approximately 2-3GB and may take several minutes depending on your internet connection.

Subsequent generations will be instant as models are cached locally.

🔒

Installation Location

If you install MyVoice to the Program Files directory, you may need to run the application as Administrator to allow proper file access and model storage.

We recommend installing to a user-accessible location like Documents or a custom folder to avoid permission issues.

Quick Start Overview

Get familiar with the interface at a glance

Main Window

Your command center for text-to-speech

Core Controls

Text Input: Type or paste the text you want to speak
Voice Selector: Choose from bundled, designed, or cloned voices
Emotion Presets: Quick buttons for Neutral, Happy, Sad, Angry, Flirtatious (F1-F5)
Generate: Create speech from your text
Replay: Replay the last generated audio
Quick Speak: Access saved phrases
Clear: Clears text (or appends apostrophe ' for Quick Speak)

Settings: Audio

Configure your audio devices and virtual microphone

Audio Configuration

Monitor Device: Your speakers/headphones for hearing playback
Virtual Device: The virtual microphone for routing to other apps
Dual Routing: When enabled, audio plays through both devices simultaneously

VB-Cable Required: To use MyVoice with Discord, Zoom, or other voice apps, install VB-Cable and select "CABLE Input (VB-Audio Virtual C)" as your Virtual Device.

Settings: Voices

Manage your voice library

Voice Library

Bundled Voices: 9 pre-trained voices with full emotion support
Designed Voices: Custom voices created from text descriptions
Cloned Voices: Voices created from audio samples

Select the Voices tab in Settings and click Design Voice to open Voice Design Studio, or use Delete to remove custom voices.

Settings: TTS

Fine-tune text-to-speech behavior

TTS Parameters

Advanced settings for controlling speech generation quality and performance. Includes model size switching to balance between quality and speed. Most users can leave these at default values.

Settings: Interface

Customize the application appearance

Interface Options

Light/Dark Theme: Choose your preferred color scheme
Always on Top: Keep MyVoice visible above other windows
Window Transparency: Adjust window opacity for overlay use
Log Level: Control diagnostic logging detail

Settings: Quick Speak

Save phrases for instant playback

Quick Speak Profiles

Profiles: Create multiple phrase collections for different contexts
Entries: Add common phrases with labels for easy identification
Append Mode: End text with apostrophe (') then select Quick Speak to append the phrase

Example: Type "Hello" and end with apostrophe, then select "How are you?" from Quick Speak. Result: "Hello How are you?"

Voice Design Studio

Create custom voices from descriptions or audio samples

1

Imagine Your Voice

Describe the voice you want to create using natural language. Be specific about characteristics like age, gender, tone, and personality.

Example descriptions:

"A warm, friendly female voice in her 30s with a slight British accent"
"A deep, authoritative male narrator voice"
"A cheerful, energetic young voice perfect for gaming"

2

Generate Emotion Samples

Generate sample audio for each emotion preset. Listen to how your voice sounds with different emotional expressions.

You can also upload custom audio samples if you have specific reference recordings.

3

Refine Your Voice

Fine-tune your voice with additional adjustments. Preview different variations and save your favorite.

Once satisfied, your custom voice will be available in the voice selector for all future speech generation.

Feature Guide