Function List

**Video Generation**

[**Image to Video**](https://platform.vidu.cn/docs/image-to-video)

**Generate a video from an input image and prompt description** - Freely select the video's aspect ratio, duration, resolution, and motion amplitude - Support for adding background music - Support for selecting multiple generation modes

[**Reference to Video**](https://vidu-platform-next.dev.shu.team/docs/reference-to-video)

**Generate a video that references the subject or scene of input images** - Support for uploading multiple images - Freely select the video's aspect ratio, duration, resolution, and motion amplitude - Support for adding background music - Support for selecting multiple generation modes

[**Start-End Frame to Video**](https://vidu-platform-next.dev.shu.team/docs/start-end-to-video)

**Input a start frame image and an end frame image, and generate a video based on the prompt** - Freely select the video's aspect ratio, duration, resolution, and motion amplitude - Support for adding background music - Support for selecting multiple generation modes

[**Text to Video**](https://platform.vidu.cn/docs/text-to-video)

**Generate a video directly from a prompt describing the scene** - Support for long text descriptions - Freely select the video's aspect ratio, duration, resolution, and motion amplitude - Support for adding background music - Support for selecting multiple generation modes

**Real-Time Interactive Digital Human**

[**Avatar Real-Time Edition**](https://platform.vidu.cn/vidu-stream/doc/s2-avatar/realtime/introduction)

Input an image, persona, and user audio/video for real-time conversation Real-time interaction where the digital human converses with the user in real time, supports interruption, and has bidirectional perception. Vidu S provides the full pipeline from speech to rendering

[**Avatar Component Edition**](https://platform.vidu.cn/vidu-stream/doc/s2-avatar/component/introduction)

Generate real-time video by referencing digital human audio + action text Real-time interaction — you already have your own RTC channel and real-time voice capability; Vidu S only needs to render the digital human and stream it to your channel

[**Avatar Non-Real-Time Edition**](https://platform.vidu.cn/vidu-stream/doc/s2-avatar/async/introduction)

Input a digital human image + digital human audio + action text to generate a video file Asynchronous generation — one character image + one audio clip (or text) generates a speaking video with aligned lip sync and movement

[**Editing Edition**](https://platform.vidu.cn/vidu-stream/doc/s2-editing/quick-start)

Input a user audio/video stream + replacement image to generate a real-time video Streaming video editing: continuously receives a video stream and outputs the edited result in real time, with the ability to switch the reference image or scene mid-stream without interrupting the stream

**Image Generation**

[**Reference to Image**](https://platform.vidu.cn/docs/reference-to-image)

**Generate an image that references the subject or scene of input images** - Support for uploading multiple images - Image characters, objects, and scenes remain highly consistent throughout - Freely select the image's aspect ratio and resolution, with support for 2K and 4K

[**Text to Image**](https://platform.vidu.cn/docs/reference-to-image)

**Generate an image directly from a prompt describing the scene** - Support for long text descriptions - Freely select the image's aspect ratio and resolution, with support for 2K and 4K

[**Image Editing**](https://platform.vidu.cn/docs/reference-to-image)

**Reference an original image and a reference image to replace elements in the original image** - Support for uploading multiple images - Support for replacing elements in the original image via text description - Freely select the image's aspect ratio and resolution, with support for 2K and 4K

**Audio Generation**

[**Text to Audio**](https://platform.vidu.cn/docs/text-to-audio)

**Generate audio directly from a prompt** - Support for generating creative sound effects or various types of background music - Freely select the audio duration

[**Controllable Text to Sound Effects**](https://platform.vidu.cn/docs/timing-to-audio)

**Generate controllable, time-sequenced sound effects directly from a prompt** - Use a timeline to flexibly control the order and duration of multiple sound effects - Support for overlapping multiple sound effects - Freely select the audio duration

[Text to Speech](https://platform.vidu.cn/docs/text-to-speech)

**Generate speech from text** - Control speech rate, pauses, volume, and polyphonic character pronunciation - Freely adjust speech emotion, such as happy, sad, or angry - Support for 30+ languages and 300+ voice options

[Voice Cloning](https://platform.vidu.cn/docs/voice-clone)

**Upload audio to clone a voice** - Invoke the cloned voice for speech synthesis at any time - Freely adjust speech emotion, such as happy, sad, or angry

**Template Usage**

[**Scene Templates**](https://platform.vidu.cn/docs/template)

**Input an image and apply an effects template with one click to generate a video** - Covers hundreds of creative templates: French kiss, journey through the universe series, and more

[**Template-Based Compilations**](https://platform.vidu.cn/docs/template-story)

**Input an image and apply a long-video template with one click to generate a compiled video** - The resulting compilation has a coherent storyline and diverse effects

**Additional Capabilities**

[**Lip Sync**](https://platform.vidu.cn/docs/lip-sync)

**Make a character in a video lip-sync to a specified text** - Freely select from hundreds of voices and languages - Support for long text input - Freely adjust volume and speech rate

[Digital Human](https://platform.vidu.cn/docs/digital-human)

**Generate a digital human video for a character in an image** - Enable the digital human to deliver speeches, broadcasts, and other content - Support for speech synthesis using 300+ voices or cloned voices

[**Video Extension**](https://platform.vidu.cn/docs/video-extension)

**Extend the duration of a video** - Support for extending by 1 to 7 seconds

[Smart Multi-Frame](https://platform.vidu.cn/docs/multi-frame)

**Input multiple keyframes to continue a video** - Use keyframes to generate high-quality videos with a coherent plot and consistent characters - Freely control the duration between frames, selectable from 2 to 7 seconds - Each frame can use a prompt to control the visual effect

[**Recommended Prompts**](https://platform.vidu.cn/docs/prompt-rec)

**Input an image to generate corresponding recommended prompts** - Support for selecting the prompt type and quantity - Support for selecting the video resolution

[**Smart Ultra HD - Premium**](https://platform.vidu.cn/docs/upscale-pro)

**Enhance the resolution of a video** - Freely select from multiple resolutions: 1080p, 2K, 4K, 8K