AI Seed Audio 1.0

INPUT

Prompt

Write the speech, scene, or audio you want. Reference uploaded clips as @Audio1, @Audio2, and @Audio3.

Length

Set the target output length in seconds.

1s 120s

Voice(s) to use

Let's you select preset voices. Leave as "Custom" to follow the voices described in your prompt or reference audio clips

Custom
Follows prompt and reference audios Select Preset
Select a voice from a preset list Voice ID
Use a SeedAudio Voice ID

Preset Language

Filter the preset voice list.

English
Default Chinese Indian

Preset Voice

Choose a provider preset voice. Leave Voice(s) to use on the default option when using @Audio references or prompt-described speakers.

Custom Voice ID

Paste a cloned voice ID accepted by fal for SeedAudio. This will be sent as the provider voice value.

Reference Audio

Optional. Upload or record up to 3 clips, then refer to them in the prompt as @Audio1, @Audio2, and @Audio3.

0/3

Reference Image Upgrade

Optional image guidance. Use either an image or audio references, not both.

0/1

Output Format Upgrade

MP3 is easiest to share. Other formats are available when you need them.

MP3
Default WAV
Uncompressed OGG Opus
Efficient PCM
Raw

Sample Rate Upgrade

Higher sample rates preserve more detail for editing and mastering.

44.1 kHz
Best for audio releases 48 kHz
Best for video and general use

Speed Upgrade

Adjust how fast the voice speaks.

Slower Faster

Volume Upgrade

Adjust the output volume.

Quieter Louder

Pitch Upgrade

Raise or lower the voice pitch.

Lower Higher

OUTPUT

Queued...

Your request will begin shortly...

Tip: You can safely leave this page and whenever the job is ready, come back and you will find the job in the "My Creations" section.

We did not forget about you. This step can sometimes take some time. But if you run a second request right after the first one completes, it is likely you will not have to wait as long since the model would be fully set up for use.

Running...