Seed Audio 1.0
All-in-one audio generation: dialogue, music, ambience, and sound effects in one pass, with voice cloning from up to 3 reference clips (under 30s each). The model field picks the variant: multilingual (default, 20 auto-detected languages, in-prompt timestamp control) or base (English/Chinese with the preset voice library).
Examples
representative generations · starting points to seed the playground
No speech. A thunderstorm approaching: distant rumble building to a close crack, heavy rain, wind through trees, 20 seconds.
[breathless, ecstatic] We did it — we actually DID it! I can't believe it, can you believe it?!
[grave, cinematic] For a thousand years, the gate remained sealed. Tonight... it opens.
Comedy UGC radio scene with THREE young women, phone-selfie vibe, close mic, natural. GIRL1 super excited showing off a thing she just bought; GIRL2 flat, deadpan, totally unbothered (the joke); GIRL3 super hyped, matches GIRL1's energy. THREE clearly DISTINCT female voices — different pitch, timbre, energy. / [SFX: room ambience, a bag rustling, something plonked on a table] / GIRL1 (super excited, fast): "Okay okay LOOK. Look what I just got, you HAVE to see this—" / GIRL3 (hyping, gasps): "Wait WHAT — no shut UP, where'd you get that?!" / GIRL1: "RIGHT?? And watch — watch what it does—" / [SFX: a soft click / little mechanism] / GIRL3 (screaming): "STOP. STOP IT. That's insane, I need one—" / GIRL2 (flat, deadpan): "...it's a cup." / GIRL1 (offended): "It is NOT just a cup—" / GIRL3: "Bestie it literally GLOWS, are you blind—" / GIRL2 (dry): "...it's a glowing cup. Cool." / GIRL1 (giddy): "It's under thirty bucks and I bought three — one each—" / GIRL3 (delighted): "Oh my god, order mine right now—" / GIRL3 (realizing): "Wait... you already bought mine?" / GIRL2 (deadpan): "It's THIRTY dollars. Are you freaking INSANE?" / [SFX: a beat of dead silence] / GIRL2 (caving, flat): "...Fine. Put me down for one." / GIRL1 and GIRL3 (screaming together): "YAASSS—" / [SFX: excited clatter, clip cuts off]
High-tension radio-drama "Talk Me Down": an air-traffic controller calmly guides a passenger who has never flown to land a plane. We are INSIDE the cockpit (close, alarms all around); the controller comes through the RADIO (filtered, tinny, faint static, unnervingly calm). Two contrasting acoustic spaces. / [SFX: cockpit alarms blaring, wind buffeting, engine whine] / PASSENGER (panicking, close): "The pilot's OUT — I don't know how to fly this! Somebody talk to me!" / CONTROLLER (through radio, filtered, calm): "I've got you. Listen to my voice. Altitude — the number, top left." / PASSENGER (shaky): "Uh — three thousand? It's dropping, it's DROPPING—" / CONTROLLER: "That's fine. Ease the yoke toward you. Gently. Like pulling a drawer." / [SFX: engine strain, a warning beep speeding up] / PASSENGER: "Why is it beeping FASTER?!" / CONTROLLER: "Ignore the beep. Eyes on the horizon. Wings level. You're doing it." / [SFX: a beat, the wind smoothing out] / PASSENGER (breath): "...it's... it's levelling out." / CONTROLLER (first hint of warmth): "Good. Now you and I are gonna land a plane. Ready?" / PASSENGER (tiny): "...not really." / CONTROLLER: "Neither was I, my first time. Throttle back, nice and slow." / [SFX: landing gear whirs down, tyres screech on the runway, roll-out rumble]
Late-night ghost-story radio piece, single narrator, testing extreme DYNAMIC RANGE. Intimate ASMR whisper, extremely close and breathy, slow, building dread and getting quieter and quieter… then ONE sudden very loud sharp scream, then instant total silence. Creepy nighttime ambience. / [SFX: dead-quiet room at night, a faint clock ticking, low wind] / NARRATOR (extreme close whisper, breathy, slow): "You know that feeling... when you're home alone... and the floor creaks upstairs?" / [SFX: a single soft creak from above] / NARRATOR (softer): "You tell yourself it's the house settling. It's always the house settling." / [SFX: the ticking slows, near total silence] / NARRATOR (barely audible): "But last night... I live on the top floor. There is no upstairs." / [SFX: a long three seconds of dead silence] / NARRATOR (whisper, trembling): "And the creaking... is getting... closer..." / [SFX: a floorboard groans right behind you] / NARRATOR (sudden, very loud, sharp scream): "WHO'S THERE?!" / [SFX: a violent bang, then instant total silence] / NARRATOR (back to a tiny calm whisper): "...goodnight."
Per-model pricing
Transparent, usage-based rates for every ByteDance model.
Seedance 2.0
Video
Image to Video
Reference to Video
Text to Video
Seedance 2.0 Fast
Video
Image to Video
Reference to Video
Text to Video
Seedance 2.0 Mini
Video
Image to Video
Reference to Video
Text to Video
Seedream 5.0 Pro Image Editing
Image
Image to Image
Text to Image
Seedream 5.0 Lite Image Editing
Image
Image to Image
Text to Image
Seedream 4.5
Image • Text to Image
Seedream 4.0 Image Editing
Image
Image to Image
Text to Image
Seed Audio 1.0Current
Audio • Text to Audio