Audio Nodes

Audio nodes add sound to your creative work. Generate background music, voiceovers, sound effects, and even clone voices.

What Are Audio Nodes?

An Audio node using MusicGen for AI music generation.

Audio nodes generate sound using AI. They can:

• 🎵 Generate music — Background tracks, jingles, songs
• 🗣️ Create speech — Voiceovers, narration
• 🎙️ Clone voices — Replicate a voice from a sample
• 🔊 Make sounds — Sound effects, ambient audio

Adding an Audio Node

1. Open the node sidebar on the left
2. Find Audio in the list
3. Drag it onto your canvas

Audio Node Inputs

Audio Models

Music Generation

Speech Generation

Voice Tools

Music Generation

Generate original music from text descriptions.

Workflow

[Text Node] → [Audio Node (Music)]
Music description → Generated music

Example

1. Add a Text Node with music description:

   Upbeat electronic music, energetic and modern,
   suitable for tech product video,
   driving beat, synth melodies
   

2. Add an Audio Node
3. Select a music model (e.g., Music 1.5)
4. Connect and run

Music Prompt Tips

Include these elements:

• Genre — "lo-fi hip hop," "cinematic orchestral"
• Mood — "uplifting," "mysterious," "relaxing"
• Instruments — "piano," "synths," "strings"
• Tempo — "slow," "upbeat," "moderate"
• Purpose — "background music," "intro jingle"

See Audio Prompting for detailed guidance.

Speech Generation (Text-to-Speech)

Convert text into spoken audio.

Workflow

[Text Node] → [Audio Node (Speech)]
Your script → Spoken audio

Example

1. Add a Text Node with your script:

   Welcome to our product demonstration. 
   Today, we'll show you how easy it is 
   to create amazing content with AI.
   

2. Add an Audio Node
3. Select Speech-02 Turbo or Speech-02 HD
4. Connect and run

Speech Settings

Voice Cloning

Replicate a specific voice from an audio sample.

Workflow

[Upload Node (Audio)] → [Audio Node (Voice Clone)]
Voice sample → Cloned speech

Steps

1. Upload a voice sample

   • Add an Upload Node
   • Upload a clear audio file (10-30 seconds)
   • Single speaker, no background noise

2. Add Audio Node

   • Select XTTS-v2 or Voice Cloning model
   • Connect the audio sample

3. Add your script

   • Text Node with what you want said
   • Connect to the Audio Node

4. Generate

   • Run to create speech in the cloned voice

Voice Sample Requirements

Audio Node Settings

Duration

Control audio length:

Quality Settings

Viewing and Downloading

When generation completes:

1. Audio player appears in the node
2. Click play to preview
3. Download button to save locally
4. Audio is auto-saved to your Gallery

Combining Audio with Video

Method 1: In Armox

Some video models accept audio input:

[Audio Node] → [Video Node]
Generated music → Video with audio

Method 2: External Editor

1. Generate video in Armox
2. Generate audio in Armox
3. Download both
4. Combine in video editor (Premiere, Final Cut, etc.)

Tips for Matching

• Generate audio after you know video duration
• Match audio mood to video content
• Consider generating music slightly longer than needed

Tips for Better Audio

Music Tips

• Be specific about genre and mood
• Include instrument preferences
• Specify tempo if important
• Mention the use case (video, podcast, etc.)

Speech Tips

• Write natural, conversational scripts
• Include punctuation for natural pauses
• Test with Turbo model first
• Use HD for final production

Voice Cloning Tips

• Use high-quality source audio
• Longer samples = better cloning
• Keep scripts natural to the voice
• Test with short phrases first

Common Issues

Music Doesn't Match Description

• Be more specific about genre
• Include more mood keywords
• Try a different model
• Simplify the description

Speech Sounds Unnatural

• Add punctuation for pauses
• Shorten long sentences
• Use more conversational language
• Try a different voice

Voice Clone Sounds Wrong

• Improve source audio quality
• Use a longer sample
• Ensure single speaker in sample
• Remove background noise from sample

Use Case Examples

YouTube Intro Music

Energetic electronic intro music,
10 seconds, catchy hook,
modern and professional,
suitable for tech channel

Podcast Voiceover

Welcome to The Creative Hour, 
where we explore the intersection 
of technology and creativity. 
I'm your host, and today we're diving into...

Product Video Background

Corporate background music,
uplifting and professional,
moderate tempo, subtle and non-distracting,
suitable for product demonstration video

Meditation Audio

Peaceful ambient music,
very slow and calming,
soft pads and gentle nature sounds,
suitable for meditation or relaxation

Next Steps

• Audio Prompting — Master audio prompts
• Audio & Music Workflow — Complete audio creation
• Video Nodes — Combine audio with video