Davinci Magihuman

Name: Davinci Magihuman
Author: Nodespell Team

Official

Talking-human video generation from an image, prompt, and optional audio.

Nodespell AI

AI / Video / Fal Ai

Talking-human video generation from an image, prompt, and optional audio.

Davinci Magihuman is useful when a still character or person image should become an expressive talking video. It exposes prompt, image, optional audio, duration, resolution, steps, guidance scale, safety, and seed.

Use it for avatar-style performance clips. Use lip-sync nodes when you already have a video and only need mouth movement matched to new audio.

Inputs (3)

Prompt

String

Text prompt describing the desired video content.

RequiredMulti InputMin: 0Max: 100

Image

String

Reference image URL for image-to-video generation.

Required

Audio

String

Optional driving audio URL for lipsync mode.

Parameters (7)

Prompt

String

Text prompt describing the desired video content.

Required

Default:

Duration

Number

Duration of the generated video in seconds.

Default: 5

Resolution

String

Output resolution. Options: 256p, 540p, 720p, 1080p. '1080p' runs base + super-resolution pipeline.

Default: 256p

Inference Steps

Number

Number of denoising steps. Defaults vary by resolution (8 for 256p, 32 for higher resolutions).

Default: 8

Guidance Scale

Number

Classifier-free guidance scale.

Default: 5

Safety Checker

Boolean

If enabled, runs safety checks on the prompt and input image.

Default: true

Seed

Number

Random seed for reproducibility. Set -1 to omit and use automatic randomness.

Outputs (1)

Output

Inferred

Generated video output.

Nodespell Team

Creator profile

Type

Node

Status

Official

Package

Nodespell AI

Keywords (7)

Video GenerationLip SyncMotion TransferAudio EnhancementConditional GenerationResolution Control

Use in Workflow