← Back to all generators

zsxkib/multitalk

Audio-driven multi-person conversational video generation - Upload audio files and a reference image to create realistic conversations between multiple people

Capabilities

Reference ImagesSeed

Cost

Community model (estimated from hardware time)

Input Parameters

first_audiorequiredstring

First audio file for driving the conversation

imagerequiredstring

Reference image containing the person(s) for video generation

num_framesinteger

Number of frames to generate (automatically adjusted to nearest valid value of form 4n+1, e.g., 81, 181)

Default: 81min: 25, max: 201
promptstring

Text prompt describing the desired interaction or conversation scenario

Default: "A smiling man and woman wearing headphones sit in front of microphones, appearing to host a podcast."
sampling_stepsinteger

Number of sampling steps (higher = better quality, lower = faster)

Default: 40min: 2, max: 100
second_audiostring

Second audio file for multi-person conversation (optional)

seedinteger

Random seed for reproducible results

turboboolean

Enable turbo mode optimizations (adjusts thresholds and guidance scales for speed)

Default: true
Version: 0bd2390c4061Updated: 7/25/20263.3K runs