Create voice or scene audio by combining a prompt with optional voice, audio, or image references.
Write the Audio Prompt
Describe the speech, speakers, mood, language, scene, background sound, or audio texture you want to generate.
Choose Voice Guidance
Use Auto, select a preset multilingual voice, or add reference audio when you need stronger voice-style direction.
Add Optional References
Upload a reference image or short reference audio clip to give Seed Audio more context for the intended result.
Generate and Download
Create the audio, review the result, refine your prompt or references if needed, and download the final file.
Write the Audio Prompt
Describe the speech, speakers, mood, language, scene, background sound, or audio texture you want to generate.
Choose Voice Guidance
Use Auto, select a preset multilingual voice, or add reference audio when you need stronger voice-style direction.
Add Optional References
Upload a reference image or short reference audio clip to give Seed Audio more context for the intended result.
Generate and Download
Create the audio, review the result, refine your prompt or references if needed, and download the final file.
Audio production often requires separate tools for voice generation, reference matching, ambience, sound design, and editing.

Seed Audio supports prompt-led generation with optional preset voices, image guidance, and short reference audio for more directed results.

Use Seed Audio when you need more flexible input control than a basic text-to-speech workflow can provide.