AI Avatar & Talking Head Videos

Create AI avatars and talking head videos via inference.sh CLI.

Quick Start

bash
1curl -fsSL https://cli.inference.sh | sh && infsh login
2
3# Create avatar video from image + audio
4infsh app run bytedance/omnihuman-1-5 --input '{
5  "image_url": "https://portrait.jpg",
6  "audio_url": "https://speech.mp3"
7}'

Available Models

Search Avatar Apps

bash
1infsh app list --search "omnihuman"
2infsh app list --search "lipsync"
3infsh app list --search "fabric"

Examples

OmniHuman 1.5 (Multi-Character)

bash
1infsh app run bytedance/omnihuman-1-5 --input '{
2  "image_url": "https://portrait.jpg",
3  "audio_url": "https://speech.mp3"
4}'

Supports specifying which character to drive in multi-person images.

Fabric 1.0 (Image Talks)

bash
1infsh app run falai/fabric-1-0 --input '{
2  "image_url": "https://face.jpg",
3  "audio_url": "https://audio.mp3"
4}'

PixVerse Lipsync

bash
1infsh app run falai/pixverse-lipsync --input '{
2  "image_url": "https://portrait.jpg",
3  "audio_url": "https://speech.mp3"
4}'

Generates highly realistic lipsync from any audio.

Full Workflow: TTS + Avatar

bash
1# 1. Generate speech from text
2infsh app run infsh/kokoro-tts --input '{
3  "text": "Welcome to our product demo. Today I will show you..."
4}' > speech.json
5
6# 2. Create avatar video with the speech
7infsh app run bytedance/omnihuman-1-5 --input '{
8  "image_url": "https://presenter-photo.jpg",
9  "audio_url": "<audio-url-from-step-1>"
10}'

Full Workflow: Dub Video in Another Language

bash
1# 1. Transcribe original video
2infsh app run infsh/fast-whisper-large-v3 --input '{"audio_url": "https://video.mp4"}' > transcript.json
3
4# 2. Translate text (manually or with an LLM)
5
6# 3. Generate speech in new language
7infsh app run infsh/kokoro-tts --input '{"text": "<translated-text>"}' > new_speech.json
8
9# 4. Lipsync the original video with new audio
10infsh app run infsh/latentsync-1-6 --input '{
11  "video_url": "https://original-video.mp4",
12  "audio_url": "<new-audio-url>"
13}'

Use Cases

• Marketing: Product demos with AI presenter
• Education: Course videos, explainers
• Localization: Dub content in multiple languages
• Social Media: Consistent virtual influencer
• Corporate: Training videos, announcements

Tips

• Use high-quality portrait photos (front-facing, good lighting)
• Audio should be clear with minimal background noise
• OmniHuman 1.5 supports multiple people in one image
• LatentSync is best for syncing existing videos to new audio

Related Skills

bash
1# Full platform skill (all 150+ apps)
2npx skills add inferencesh/skills@inference-sh
3
4# Text-to-speech (generate audio for avatars)
5npx skills add inferencesh/skills@text-to-speech
6
7# Speech-to-text (transcribe for dubbing)
8npx skills add inferencesh/skills@speech-to-text
9
10# Video generation
11npx skills add inferencesh/skills@ai-video-generation
12
13# Image generation (create avatar images)
14npx skills add inferencesh/skills@ai-image-generation

Browse all video apps: infsh app list --category video

Documentation

• Running Apps - How to run apps via CLI
• Content Pipeline Example - Building media workflows
• Streaming Results - Real-time progress updates