Home/Models/p-video-avatar

P Video Avatar

Video
replicate
video

No description available.

Ready
Run the model to preview output.

README

# P Video Avatar ## Overview No README was provided for this model yet. ## Model Metadata - Provider: replicate - Category: Video - Output Type: video - Slug: p-video-avatar ## Notes Use the Playground tab above to test prompts and parameters.

Parameters

Model input reference derived from preset schema.

NameTypeRequiredDefaultDescription
seedintegerNonullRandom seed. Set for reproducible generation.
audioaudioNonullOptional uploaded audio to drive avatar speech. If provided, this is used instead of voice_script and voice settings.
imageimageYesnullInput image (first frame). Supports jpg, jpeg, png, webp.
no_opbooleanNofalseHealth check mode - returns status without inference.
voiceselectNo"Zephyr (Female)"Voice to use when generating speech from voice_script.
resolutionselectNo"720p"Resolution of the video.
video_promptstringNo"The person is talking."Optional visual prompt describing how the person should appear or behave while speaking.
voice_promptstringNo"Say the following."Optional style instructions for how to speak voice_script, such as tone, pacing, accent, or emotion. These instructions are not spoken.
voice_scriptstringNo""Exact words the avatar should say. Required when no audio file is uploaded.
voice_languageselectNo"English (US)"Language/accent target for generated speech.
negative_promptstringNo""Disabled if empty.Mention what you do NOT want in the video, e.g. "subtitles, text, blurry, low quality, frames, watermark, titles, scene change". We recommend using multiple keywords at once.
disable_safety_filterbooleanNotrueDisable safety filter for prompts and input image. When disabled, prompts are not checked for unsafe content before generation.
strength_negative_promptnumberNo0.5Strength of the Negative Prompt. Optimal value can differ for different video lengths (Experimental Feature)
disable_prompt_upsamplingbooleanNofalseWhen true, skip automatic enhancement of the visual video prompt and use video_prompt directly.