mirror of
https://github.com/jamiepine/voicebox.git
synced 2026-10-04 01:25:18 -07:00
66 lines
1.3 KiB
Plaintext
66 lines
1.3 KiB
Plaintext
---
|
|
title: "Generating Speech"
|
|
description: "Generate high-quality speech from text"
|
|
---
|
|
|
|
## Basic Generation
|
|
|
|
<Steps>
|
|
<Step title="Select Profile">
|
|
Choose a voice profile from the dropdown
|
|
</Step>
|
|
<Step title="Enter Text">
|
|
Type or paste your text
|
|
</Step>
|
|
<Step title="Generate">
|
|
Click **Generate** and wait a few seconds
|
|
</Step>
|
|
<Step title="Play & Export">
|
|
Preview and download the result
|
|
</Step>
|
|
</Steps>
|
|
|
|
## Text Formatting Tips
|
|
|
|
The way you format text affects the output quality.
|
|
|
|
### Punctuation
|
|
|
|
Use proper punctuation for natural pauses:
|
|
|
|
```
|
|
Good: "Hello! How are you today? I'm doing great."
|
|
Bad: "Hello how are you today Im doing great"
|
|
```
|
|
|
|
### Emphasis
|
|
|
|
Use formatting to suggest emphasis:
|
|
|
|
```
|
|
- ALL CAPS for louder/emphasized: "That was AMAZING!"
|
|
- Italics for subtle emphasis: "I *really* enjoyed that"
|
|
- Bold for strong emphasis: "This is **very** important"
|
|
```
|
|
|
|
<Callout type="info">
|
|
The model interprets these hints but results may vary.
|
|
</Callout>
|
|
|
|
## Advanced Features
|
|
|
|
### Batch Generation
|
|
|
|
For long-form content, split into smaller chunks for better control and faster processing.
|
|
|
|
### Voice Caching
|
|
|
|
Voicebox caches voice prompts for faster re-generation with the same profile.
|
|
|
|
## Coming Soon
|
|
|
|
- Real-time streaming
|
|
- Word-level timing control
|
|
- Emotion and style controls
|
|
- SSML support
|