skills.rest
Aadi-110i
Aadi-110iCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

talking-head-production

Generate talking head videos with AI avatars, lipsync, and built-in TTS via inference.sh CLI.

Updated Jul 28, 2026
Search Tags:#text-to-speech#video generation#inference.sh#lipsync#talking head#ai avatar#omnihuman
One-click install
npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill talking-head-production-aadi-110i
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: talking-head-production
Source: https://github.com/Aadi-110i/PEP-PROJECT/tree/main/skills/talking-head-production
Command: npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill talking-head-production-aadi-110i

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Producing spokesperson, course, or social media videos traditionally requires cameras, actors, and editing. This Skill generates talking head videos from a single portrait image and a script using AI avatar models with lipsync and voiceover. ## Core Features & Use Cases - Text-to-Avatar Video: P-Video-Avatar converts a portrait image plus a voice script into a lipsynced video with 30 built-in voices across 10 languages. - Model Selection Guidance: Compares P-Video-Avatar, OmniHuman 1.5/1.0, Fabric 1.0, and PixVerse Lipsync by speed, cost, and use case. - Production Workflows: Covers portrait generation, custom audio input, multi-character dialogue, long-form clip stitching, multilingual content, and dubbing existing videos. - Use Case: A course creator generates a professional headshot with P-Image, then produces a 10-language presenter video series using P-Video-Avatar's built-in TTS without recording any audio. ## Quick Start Use the talking-head-production skill to create a talking head video from my portrait image with the script "Welcome to our product tour" using P-Video-Avatar.

Frequently Asked Questions about talking-head-production

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a talking head video from a photo?▼

Run P-Video-Avatar via the belt CLI with a portrait image URL and a voice_script parameter. The model generates a lipsynced video with built-in TTS, so no separate audio recording is needed.

Which AI avatar model is best for talking head videos?▼

P-Video-Avatar is recommended for most cases at roughly 1.83 seconds per second of video and $0.025 per second with built-in TTS. OmniHuman 1.5 suits multi-character scenes, while Fabric and PixVerse Lipsync cover alternative lipsync needs.

What portrait image requirements does AI lipsync need?▼

The portrait must be center-framed, head and shoulders, with eyes looking at the camera and a neutral expression. The face region should be at least 512x512 pixels, ideally 1024x1024 or higher, with no sunglasses or obstructions.

Can I use my own audio instead of built-in TTS?▼

Yes, pass an audio file URL via the audio parameter for P-Video-Avatar or audio_url for OmniHuman. Use clean audio at 44.1kHz or 48kHz in MP3 128kbps+ or WAV format for accurate lipsync.

How do I make talking head videos longer than 60 seconds?▼

Split the script into segments and generate separate clips with the same portrait using the --no-wait flag. Then merge the clips with the infsh/media-merger app to produce the final long-form video.

Why does my AI avatar video have bad lipsync?▼

Poor lipsync usually comes from low-resolution portraits, side-angle faces, or noisy audio. Use a frontal 1024x1024+ portrait and clean audio, or switch to P-Video-Avatar's built-in TTS to eliminate audio quality issues.

Related Skills

Explore More Content & Communication Skills→
genmedia-labsgenmedia-labs
12

lipsync

Generates lip-synced videos from audio tracks and source media via the RunComfy CLI.

Official
Intermediate
fdsfds3232fdsfds3232

clip-hand-skill

Transform long-form videos into captioned short clips for social media.

Community
Advanced
compozycompozy
182

hyperframes-media

Generate TTS audio, word-timestamp transcripts, and transparent overlays for HyperFrames.

Official
Advanced
inference-shinference-sh
688

ai-avatar-video

Generates AI avatar and talking head videos via the inference.sh CLI.

Official
Intermediate