Seed Audio 1.0 is ByteDance Seed's all-in-one audio generation model for creating complete sound scenes. Use text, image, or audio context to guide multi-speaker dialogue, emotional delivery, native accents, ambience, background music, and foley-style effects.
Seed Audio 1.0
Scene prompt preview
Prompt concept
Two speakers whisper in a rainy alley, tense strings underneath, distant traffic, footsteps, and a final metallic door slam.
Use the Seed Audio 1.0 workspace below to create sound scenes from a prompt, optional reference audio, or one reference image.
Input
Prompt-first audio generation with optional controls.
Additional Settings
Customize your input with more control.
History
Your recent Seed Audio 1.0 Preview generations.
Sign in to see your generation history.
Model overview
Seed Audio 1.0 is ByteDance Seed's multimodal AI audio generation model for building complete sound scenes from text, image, and audio inputs. In one workflow, it can combine multi-speaker dialogue, expressive voices, native accents, ambience, background music, and foley-style sound effects instead of producing speech alone.
Prompts can describe speakers, timing, emotion, language, location, music direction, and sound events together. This makes the model useful for cinematic scenes, localized dialogue, ads, podcasts, games, and other projects where voice and the surrounding soundscape need to work as one.
Learn more about Seed Audio 1.0Inputs
Text, image, and up to 3 audio references
Sound layers
Dialogue, voice, ambience, music, and sound effects
Output
One coordinated audio scene up to 2 minutes
How to use
To use Seed Audio 1.0 online, write a sound-scene prompt, optionally add an audio or image reference, choose output settings, then generate and review the audio in the SeedAudio.co workspace.
Describe the characters, language, emotion, location, dialogue, ambience, music direction, and sound events you want in the scene.
Use up to three audio references for voice or style direction, or one image reference to guide mood and scene context.
Pick voice behavior, output format, sample rate, speed, volume, pitch, and a max credit cap before running the generation.
Run a short draft first, listen for voice clarity and layer balance, then copy, download, or revise the result.
Prompt formula
Scene + speaker + emotion + language + ambience + BGM + sound effects + timing.
Built for real production workflows
Create precisely timed foley, original music, multilingual dialogue, and reusable authorized voices—from one production brief.
Turn a shot list into edit-ready footsteps, fabric, props, and ambience for short dramas, food videos, and AI video post-production.
Create a 12-second pure foley scene in a rainy wooden teahouse: a silk sleeve brushes the table, three leather steps approach, a door slides open, and a porcelain cup lands on wood. No dialogue or music.
Rain at the Teahouse
0:12
Produce original soundtracks, hooks, and background music for films, ads, and regional formats without hunting through stock libraries.
Compose a cinematic Silk Road night-drive cue with plucked lute, frame drum, and modern bass. Build to a memorable hook at eight seconds. Instrumental only.
Silk Road After Dark
0:16
Localize dialogue, ads, and narrated content with distinct roles, controlled emotion, and native-sounding delivery—not flat, one-voice reading.
At a harbor before dawn, create the same restrained two-character exchange in Italian and Chinese. Keep native pronunciation, soft harbor ambience, and no music.
Harbor Light · Italian
0:14
Keep an authorized character or brand voice consistent across episodes, campaigns, and markets while changing the script on demand.
Use @Audio1 as the authorized voice reference. Preserve its warm identity, measured pace, and gentle accent while reading: ‘Every new chapter begins with one clear breath.’
Authorized reference
Quiet Hour · Original
0:09Generated variation
Next Chapter · Cloned
0:07Seed Audio 1.0 vs traditional TTS
Traditional TTS is designed to turn written text into spoken voice. Seed Audio 1.0 is designed for broader audio generation, combining dialogue, performance direction, ambience, music, and sound effects in one sound-scene workflow.
| Capability | Seed Audio 1.0 | Traditional TTS (early commercial TTS and basic speech synthesis) |
|---|---|---|
| Core positioning | Scene-level audio creation | Text-to-speech |
| Generated content | Spoken dialogue + background music + sound effects + ambience, generated and mixed in one pass | Voice only (a single speech track) |
| Multiple speakers | Native support for generating coordinated multi-character dialogue, emotion, and pacing in one pass | Usually requires repeated calls with different voices, followed by manual alignment and mixing |
| Background music & sound effects | Generated with the dialogue and automatically matched to its emotion and timeline | Not included; requires separate music models, sound libraries, and a DAW |
| Input method | Text prompt / reference audio (up to 3 clips) / reference image | Mainly text + a preset voice; some systems support basic references |
| Voice cloning | Zero-shot cloning from a short uploaded reference clip (about 30 seconds or less) | Often requires fine-tuning or training, or relies on a limited voice library |
| Emotion & expressiveness control | Describe emotion, tone, accent, and pacing directly in natural language | Primarily controlled through SSML tags or limited parameters, with a narrower expressive range |
| Timeline control | Fine timestamp control over when dialogue appears | Generally not available |
| Output format | A fully mixed, non-streaming audio track up to about 2 minutes per generation | A single voice file that requires music and sound effects to be added later |
| Best suited for | Audio drama, short-form dubbing, ads, game audio, podcast cold opens, video narration, and other complete sound scenes | Audiobook reading, navigation, customer service, simple narration, and other voice-only needs |
| Workflow change | One prompt → finished audio track, substantially reducing post-production mixing | Generate speech → find music → find sound effects → align and mix manually |
Traditional TTS remains the simpler choice when you only need predictable speech. Choose Seed Audio 1.0 when the surrounding scene and sound design are part of the output.
Straightforward workflow
Move from a sound idea to a complete scene direction: define characters, emotion, location, dialogue, music, ambience, and effects in one prompt.
Start with characters, emotion, location, timing, dialogue, musical mood, and the effects that should exist in the scene.
Seed Audio 1.0 is designed to synthesize dialogue, emotional tone, native accents, ambience, BGM, and distinct sound effects together.
Shape audio directions for short films, ads, podcasts, games, learning content, and other projects that need coherent sound scenes quickly.
Seed Audio 1.0 technology
Seed Audio 1.0 is positioned for complete audio scenes: multi-character dialogue, emotion, tone, accents, ambience beds, BGM, and foley in a single creative pass.
Multi-speaker
voice continuity for longer generated scenes
Text / image / audio
multimodal prompting for audio creation
Compose multiple sound layers at once instead of stitching voice, music, ambience, and effects in separate tools.
Guide tone, emotional delivery, dialect, and native-sounding accents while keeping recurring voices recognizable across contexts.
Generate environmental beds, background music, room tone, weather, crowds, or distant city texture alongside the dialogue.
Seed Audio 1.0 is built for longer-form sound scenes, including session-length generation suitable for dialogue, ambience, and music-backed sequences.
Creative possibilities
Seed Audio 1.0 is most interesting when a project needs more than narration: a complete acoustic scene with voices, mood, space, and events.
Draft dialogue, emotional beats, foley, ambience, and music for storyboards or pre-visualization.
Create campaign-ready sound directions for product demos, social clips, and localized ads.
Prototype ambient loops, character barks, UI sounds, and cinematic moments before a final audio pass.
Build scenario-based lessons, character conversations, and immersive explainers with spatial sound cues.
Multi-character delivery with emotional tone
Rain, traffic, rooms, crowds, and natural beds
Footsteps, impacts, doors, texture, and timing
Pricing
Subscribe for the best value, or buy credits when you need a flexible top-up. Every paid plan and credit pack includes Seed Audio 1.0 API access, with one shared credit balance across the web app and API.
See Seed Audio 1.0 pricing detailsTry Seed Audio 1.0 with free signup credits. Perfect for testing prompts and short scenes.
The best annual choice for creators with ongoing audio needs.
The strongest value for teams that expect heavy audio generation.
FAQs
Practical answers for creators who want to understand Seed Audio 1.0 and use it for AI audio generation.
Follow the model's capabilities, access status, and practical use cases for multimodal AI audio generation across dialogue, ambience, music, and sound effects.