# Seedance v1.5 Pro

Seedance v1.5 Pro moves from visual-only output to joint audio-visual creation. It generates voice, ambient sound, and video together with aligned lip movement and intonation.

- **Model ID:** `bytedance/seedance-v1.5-pro`
- **Type:** video
- **Providers:** bytedance
- **Context window:** Not applicable
- **Maximum output tokens:** 0
- **Pricing:** See the provider details below
- **Canonical page:** https://vercel.com/ai-gateway/models/seedance-v1.5-pro

## Supported parameters

- **Parameters:** `aspectRatio`, `resolution`, `duration`, `fps`
- **Operations:** `text-to-video`, `image-to-video`, `first-last-frame`
- **Input modalities:** `text`, `image`
- **Output modalities:** `video`
- **Aspect ratios:** `16:9`, `4:3`, `1:1`, `3:4`, `9:16`, `21:9`
- **Resolutions:** `720p`, `1080p`
- **Durations:** `5s`, `10s`
- **Frame rates:** `24 fps`
- **Maximum outputs per request:** Not reported
- **Maximum input images:** 2
- **Input image formats:** `jpeg`, `png`, `webp`, `bmp`, `tiff`, `gif`, `heic`, `heif`
- **Maximum input image size:** 30 MB

## Example

```ts
import { experimental_generateVideo as generateVideo } from 'ai';

const result = await generateVideo({
  model: 'bytedance/seedance-v1.5-pro',
  prompt: 'A serene mountain lake at sunrise.'
});
```

## About

Seedance v1.5 Pro shifts the Seedance line from visual generation alone to joint audio-visual creation. Released December 16, 2025, it's the first Seedance model to generate voice, sound effects, and ambient audio synchronized to video in a single inference pass. You don't run a separate text-to-speech or audio compositing step.

The audio system supports multilingual speech generation across six languages: Chinese, English, Japanese, Korean, Spanish, and Indonesian. It also covers regional dialects such as Sichuanese and Cantonese. Vocal synthesis targets prosody and intonation that track the scene. Spatial reverb in sound effects matches the visual scene's physical context. ByteDance's release cites lip movement alignment, intonation patterning, and performance rhythm synchronization as focus areas versus listed baselines. See https://console.byteplus.com/ark/region:ark+ap-southeast-1/model/detail?Id=seedance-1-5-pro for tables and comparisons.

On the video side, Seedance v1.5 Pro raises the ceiling relative to Seedance 1.0 Pro. Where 1.0 focused on motion stability, 1.5 Pro extends camera control and finishing. You get cinematic camera controls including continuous long takes and dolly zooms, color grading controls, more facial detail in close-ups, and richer dynamic motion. Output supports 480p, 720p, and 1080p resolution at 24 fps, with clips from four to 12 seconds and seven aspect ratios.

## What to consider

For audio-visual workflows, confirm that your integration layer handles the combined audio-video output format before you deploy to production. Compare rates (N/A; N/A).

## When to use

### Best For

- Character-driven video content where synchronized dialogue, lip alignment, and vocal performance are part of the creative brief
- Multilingual content production for markets requiring Chinese, English, Japanese, Korean, Spanish, or Indonesian voice, including regional dialect variants
- Cinematic short-form content where dolly zooms, long takes, and color grading go beyond typical social clip defaults
- Ambient-audio video storytelling, product demos, branded content, and explainer videos, where spatial sound effects match the scene without manual audio post-production

### Consider Alternatives When

- Audio isn't a requirement and the project is purely visual. Use Seedance 1.0 Pro for lower cost on visual-only pipelines
- Maximum generation speed and cost efficiency are the primary drivers. Use Seedance 1.0 Pro Fast
- You need a dialect or language not yet supported by the audio system. Verify support before committing

## Best for

- **Character-driven video content:** Creative briefs that include synchronized dialogue, lip alignment, and vocal performance
- **Multilingual content production:** Chinese, English, Japanese, Korean, Spanish, or Indonesian voice plus regional dialect variants
- **Cinematic short-form content:** Dolly zooms, long takes, and color grading that go beyond typical social clip defaults
- **Ambient-audio storytelling:** Product demos, branded content, and explainer videos where spatial sound effects match the scene without manual audio post-production

## Consider alternatives

- **Visual-only pipelines:** Seedance 1.0 Pro offers lower cost when audio isn't a requirement
- **Maximum speed and cost efficiency:** Seedance 1.0 Pro Fast is the primary choice when those drivers dominate
- **Unsupported languages:** Verify support before committing when you need a dialect or language not yet covered by the audio system

## Frequently asked questions

### What languages does Seedance v1.5 Pro support for voice generation?

Six languages: Chinese, English, Japanese, Korean, Spanish, and Indonesian. Regional dialect coverage includes Sichuanese and Cantonese.

### Does Seedance v1.5 Pro require a separate text-to-speech step for audio?

No. Seedance v1.5 Pro generates voice, ambient sound, and sound effects in the same inference pass as the video. You don't need an external audio pipeline.

### How does audio-visual synchronization work in Seedance v1.5 Pro?

Seedance v1.5 Pro trains to align lip movements, intonation patterns, and performance rhythm with visual content. ByteDance's release documentation reports lower audio-visual misalignment than listed baselines in its tables. See https://console.byteplus.com/ark/region:ark+ap-southeast-1/model/detail?Id=seedance-1-5-pro.

### What video specifications does Seedance v1.5 Pro support?

Resolutions of 480p, 720p, and 1080p at 24 fps, clip durations from four to 12 seconds, and seven aspect ratios: 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, and 9:21.

### How does Seedance v1.5 Pro differ from Seedance 1.0 Pro on video quality alone?

Seedance 1.5 Pro adds cinematic camera techniques (dolly zooms, long takes), color grading controls, and more facial detail in close-ups, beyond the motion stability focus of 1.0 Pro.

### Can Seedance v1.5 Pro generate ambient sound without spoken dialogue?

Yes. The audio system generates spatial sound effects and ambient audio that match the visual scene's physical environment, whether or not the scene contains speech.

## Links

- [Model page](https://vercel.com/ai-gateway/models/seedance-v1.5-pro)
- [AI Gateway documentation](https://vercel.com/docs/ai-gateway)
- [Provider model documentation](https://console.byteplus.com/ark/region:ark+ap-southeast-1/model/detail?Id=seedance-1-5-pro)
- [Provider pricing](https://docs.byteplus.com/en/docs/ModelArk/1544106)


If you encounter incorrect information, can’t find what you need, need a capability Vercel doesn’t currently offer, or have any other feedback to report, please do it by following the instructions at https://vercel.com/feedback.md.