Text-to-Audio Node

Convert text into speech or generate sound effects using AI audio models.

Overview

The Text-to-Audio node generates audio from text input. Use it for voiceovers, narration, dialogue, or sound generation in your Canvas workflows.

Inputs

Text Content (required)

Text to convert to speech
Can connect from Prompt Node
Or type directly

Outputs

Audio File

MP3 or WAV format
Natural-sounding speech
Saved to Media Library

Configuration

Voice Selection

50+ AI voices
Male/female options
Various accents
Character voices
Professional/casual styles

Voice Settings

Speed: 0.5x - 2.0x

Slower for emphasis
Faster for energetic delivery
1.0x = natural pace

Pitch: -12 to +12 semitones

Lower = deeper voice
Higher = lighter voice
Subtle adjustments recommended

Stability: 0-100%

Low = more emotion/variation
High = consistent/stable

Use Cases

Narration

Text: "In a world where..." Voice: Professional narrator Result: Documentary-style narration

Character Dialogue

Text: "Hello! How are you today?" Voice: Cheerful character Result: Character voice line

Instructions

Text: "Step 1: Open the application..." Voice: Clear instructor Result: Tutorial audio

Marketing

Text: "Discover the future of..." Voice: Professional announcer Result: Commercial voiceover

Workflow Integration

Input from:

Prompt Node
Chat Node (generated text)
Notes Node

Output to:

Voice-Over Node (add to video)
Video Stitcher (audio track)
Asset Node (save/reuse)
Timeline Editor

Voice Categories

Professional

News anchor
Corporate narrator
Documentary voice
Audiobook reader

Conversational

Friendly/casual
Warm/inviting
Natural dialogue

Character

Young/energetic
Wise/elderly
Fantasy/sci-fi
Cartoon/animated

Specialized

ASMR/whisper
Sports commentator
Radio DJ
Storyteller

Tips for Better Audio

Text Formatting:

Use punctuation for pacing
Add "..." for pauses
ALL CAPS for emphasis (use sparingly)
Break long text into sections

Voice Selection:

Match voice to content tone
Test multiple voices
Preview before generating
Save favorites

Settings:

Start with default settings
Adjust speed for content type
Subtle pitch changes work best
Higher stability for consistency

Best Practices

Writing for Speech:

Write how people speak
Use contractions (don't, can't)
Avoid complex sentences
Include natural pauses

Audio Quality:

Generate in chunks (not too long)
Use high stability for professional
Lower stability for emotional
Test different voices

Processing Time

Short text (1-2 sentences): 5-10 seconds
Paragraph: 10-20 seconds
Longer content: 20-60 seconds

Troubleshooting

Robotic sound:

Try different voice
Adjust stability setting
Rewrite text more naturally
Add punctuation for flow

Wrong pronunciation:

Spell phonetically
Use punctuation
Try different voice
Break into smaller parts

Flat delivery:

Lower stability
Add emphasis markers
Rewrite for emotion
Try character voices

Cost & Usage

Audio counted in minutes:

30 seconds = 0.5 minutes
1 minute = 1 minute
Counts toward monthly audio limit

View pricing →