ワンクリックでManusで任意のスキルを実行

avatar-video

スター5

フォーク0

更新日2026年7月6日 02:31

Generates talking avatar videos by chaining image generation (nano-banana), voice synthesis (MiniMax Speech via fal.ai), and lip-sync animation (OmniHuman 1.5 or Infinite Talk via fal.ai). Use when asked to "create a talking avatar", "generate a video with speech", "make an avatar say something", "lip-sync", or any video generation task combining portrait + voice + animation.

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

Manusで実行

ソース

barkleesanders

barkleesanders/claude-code-starter

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Manusで実行

avatar-video

Full pipeline: text script → portrait image → voice audio → lip-synced video.

/init - First-Time Setup

~/.claude/skills/avatar-video/scripts/setup.sh

This creates ~/.avatar-video/.env with your FAL_KEY. Get one at https://fal.ai → Dashboard → API Keys.

nano-banana must also be installed (see /nano-banana skill).

Quick Reference

Full Pipeline (text → video)

# Premium quality (OmniHuman 1.5) — ~$3.68 for 30s
~/.claude/skills/avatar-video/scripts/generate-avatar.sh \
  "Hi, I'm your your app assistant. I'm here to help you navigate your VA disability claim." \
  --quality premium

# Budget quality (Infinite Talk) — ~$1.88 for 30s
~/.claude/skills/avatar-video/scripts/generate-avatar.sh \
  "Welcome to our service" \
  --quality budget

Individual Steps

# 1. Generate portrait (uses nano-banana)
nano-banana "professional headshot of a woman, studio lighting, neutral background" -s 1K -o portrait

# 2. Generate voice (MiniMax Speech)
~/.claude/skills/avatar-video/scripts/minimax-tts.sh \
  "Hello, welcome to your app" \
  --voice Friendly_Person \
  --output speech.mp3

# 3a. Lip-sync with OmniHuman 1.5 (premium)
~/.claude/skills/avatar-video/scripts/omnihuman-lipsync.sh \
  --image portrait.png \
  --audio speech.mp3 \
  --output avatar-video.mp4

# 3b. Lip-sync with Infinite Talk (budget)
~/.claude/skills/avatar-video/scripts/infinitetalk-lipsync.sh \
  --image portrait.png \
  --audio speech.mp3 \
  --output avatar-video.mp4

Orchestrator Options

Flag	Default	Description
`--quality`	`premium`	`premium` (OmniHuman) or `budget` (Infinite Talk)
`--engine`	auto	`omnihuman` or `infinitetalk` (overrides --quality)
`--voice`	`Friendly_Person`	MiniMax voice ID
`--voice-speed`	`1.0`	Speech speed (0.5-2.0)
`--image`	-	Skip image gen, use existing portrait
`--audio`	-	Skip voice gen, use existing audio
`--output`	`avatar-{timestamp}`	Output filename (no extension)
`--dir`	current directory	Output directory
`--image-prompt`	auto	Custom prompt for portrait generation

MiniMax Speech Voices

Built-in voices (pass to --voice):

Voice ID	Description
`Wise_Woman`	Mature, authoritative female
`Friendly_Person`	Warm, approachable (default)
`Deep_Voice_Man`	Deep, resonant male
`Calm_Woman`	Soothing, relaxed female
`Casual_Guy`	Informal, conversational male
`Lively_Girl`	Energetic, youthful female

Custom voice cloning: provide a voice sample URL as --voice <url_to_6sec_sample>.

Engine Comparison

Feature	OmniHuman 1.5	Infinite Talk
Cost	$0.12/s (lip-sync)	$0.03/s (480p), $0.06/s (720p)
30s video	~$3.60	~$0.90-$1.80
Max length	30s audio	10 minutes
Quality	Best — full body motion, gestures	Good — solid lip-sync
Resolution	720p/1080p	480p/720p
Best for	Short premium content, ads	Long-form, bulk, prototyping

Full Pipeline Costs (30-second video)

Step	Tool	Cost
Image	nano-banana (Gemini Flash)	~$0.05
Voice	MiniMax Speech 2.6 Turbo	~$0.03
Lip-sync (budget)	Infinite Talk 720p	~$1.80
Lip-sync (premium)	OmniHuman 1.5	~$3.60
Total (budget)		~$1.88
Total (premium)		~$3.68

Cost Tracking

Every generation logs to ~/.avatar-video/costs.json. View summary:

~/.claude/skills/avatar-video/scripts/generate-avatar.sh --costs

API Keys

Service	Key	Source
MiniMax Speech, OmniHuman, Infinite Talk	`FAL_KEY`	`~/.avatar-video/.env`
nano-banana (image gen)	`GEMINI_API_KEY`	`~/.nano-banana/.env`

All fal.ai services use a single key. Get one at https://fal.ai.

Troubleshooting

"FAL_KEY not found" — Run setup: ~/.claude/skills/avatar-video/scripts/setup.sh
"Audio must be under 30s" (OmniHuman) — Use Infinite Talk for longer audio, or trim script
Upload failed — Check FAL_KEY is valid; fal.ai file upload requires auth
Queue timeout — OmniHuman can take 2-5 min; the script polls automatically
nano-banana not found — Install via /nano-banana skill init

このリポジトリの他の Skills

同じリポジトリ

printing-press-amend

barkleesanders/claude-code-starter

Amend a published CLI from one of two input sources: (1) dogfood mode mines the active Claude Code session transcript for friction (missing flags, hand- rolled API payloads, silent-null returns); (2) direct-input mode accepts user-supplied asks (rename a command, add commands or feeds, fix a named bug, optionally sniff the source site for new endpoints). Confirms scope with the user, plans + executes the fix autonomously, scrubs PII, and opens a PR against mvanhorn/printing-press-library. Two user-in-loop checkpoints: scope after capture, PR draft before open. Trigger phrases: "amend the CLI", "submit a patch", "fix what I just dogfooded", "open a PR for this CLI", "patch this CLI", "add features to my CLI", "rename this command", "add these feeds to <cli>", "sniff for new APIs in <cli>", "amend with these ideas", "use printing-press-amend", "run printing-press-amend".

2026-07-255

printing-press-import

barkleesanders/claude-code-starter

Bring a published CLI from the public library into the internal library so it's identical to a freshly-generated copy — module path reverted, manuscripts placed alongside, ready for /printing-press-polish or /printing-press-emboss. Use when the public library has a CLI you don't have locally, or to recover from a broken/lost internal copy. Trigger phrases: "import the CLI", "bring it into my library", "fetch from public library", "I don't have it locally yet".

2026-07-255

printing-press-output-review

barkleesanders/claude-code-starter

Internal sub-skill: agentic review of a printed CLI's sampled command output for plausibility issues that rule-based checks can't encode (substring-match relevance, format bugs, silent source drops, ranking failures). Invoked via the Skill tool by main printing-press SKILL.md (Phase 4.85) and printing-press-polish SKILL.md during the diagnostic loop. Not for direct user invocation — its actionable wrappers are /printing-press and /printing-press-polish.

2026-07-255

printing-press-polish

barkleesanders/claude-code-starter

Polish a generated CLI to pass verification and become publish-ready. Runs diagnostics (dogfood, verify, scorecard, go vet, gosec), automatically fixes all issues (verify failures, static-analysis findings, dead code, descriptions, README, MCP tool quality), reports the before/after delta, and offers to publish. Use after any /printing-press run, or on any CLI in $PRESS_LIBRARY/. Trigger phrases: "polish", "improve the CLI", "fix verify", "make it publish-ready", "clean up the CLI", "get this ready to ship".

2026-07-255

printing-press-reprint

barkleesanders/claude-code-starter

Regenerate an existing printed CLI from scratch under the current Printing Press, with prior research, prior novel features, and prior patches (post-publish hand-fixes) carried into the writing pipeline as reconciliation context rather than dropped on the floor. Pulls the CLI from the public library if it isn't local, recommends reuse-vs-redo of prior research based on age, then hands off to /printing-press with the right context. Use when a machine upgrade would benefit a published CLI more than manual polish. Trigger phrases: "reprint <api>", "regenerate <api>", "redo the <api> CLI", "rebuild <api> from scratch", "this CLI would benefit from a reprint".

2026-07-255

printing-press-score

barkleesanders/claude-code-starter

Score a generated CLI against the Steinberger bar, compare two CLIs side-by-side

2026-07-255

name	avatar-video
user-invocable	true
description	Generates talking avatar videos by chaining image generation (nano-banana), voice synthesis (MiniMax Speech via fal.ai), and lip-sync animation (OmniHuman 1.5 or Infinite Talk via fal.ai). Use when asked to "create a talking avatar", "generate a video with speech", "make an avatar say something", "lip-sync", or any video generation task combining portrait + voice + animation.

avatar-video

Full pipeline: text script → portrait image → voice audio → lip-synced video.

/init - First-Time Setup

~/.claude/skills/avatar-video/scripts/setup.sh

This creates ~/.avatar-video/.env with your FAL_KEY. Get one at https://fal.ai → Dashboard → API Keys.

nano-banana must also be installed (see /nano-banana skill).

Quick Reference

Full Pipeline (text → video)

# Premium quality (OmniHuman 1.5) — ~$3.68 for 30s
~/.claude/skills/avatar-video/scripts/generate-avatar.sh \
  "Hi, I'm your your app assistant. I'm here to help you navigate your VA disability claim." \
  --quality premium

# Budget quality (Infinite Talk) — ~$1.88 for 30s
~/.claude/skills/avatar-video/scripts/generate-avatar.sh \
  "Welcome to our service" \
  --quality budget

Individual Steps

# 1. Generate portrait (uses nano-banana)
nano-banana "professional headshot of a woman, studio lighting, neutral background" -s 1K -o portrait

# 2. Generate voice (MiniMax Speech)
~/.claude/skills/avatar-video/scripts/minimax-tts.sh \
  "Hello, welcome to your app" \
  --voice Friendly_Person \
  --output speech.mp3

# 3a. Lip-sync with OmniHuman 1.5 (premium)
~/.claude/skills/avatar-video/scripts/omnihuman-lipsync.sh \
  --image portrait.png \
  --audio speech.mp3 \
  --output avatar-video.mp4

# 3b. Lip-sync with Infinite Talk (budget)
~/.claude/skills/avatar-video/scripts/infinitetalk-lipsync.sh \
  --image portrait.png \
  --audio speech.mp3 \
  --output avatar-video.mp4

Orchestrator Options

Flag	Default	Description
`--quality`	`premium`	`premium` (OmniHuman) or `budget` (Infinite Talk)
`--engine`	auto	`omnihuman` or `infinitetalk` (overrides --quality)
`--voice`	`Friendly_Person`	MiniMax voice ID
`--voice-speed`	`1.0`	Speech speed (0.5-2.0)
`--image`	-	Skip image gen, use existing portrait
`--audio`	-	Skip voice gen, use existing audio
`--output`	`avatar-{timestamp}`	Output filename (no extension)
`--dir`	current directory	Output directory
`--image-prompt`	auto	Custom prompt for portrait generation

MiniMax Speech Voices

Built-in voices (pass to --voice):

Voice ID	Description
`Wise_Woman`	Mature, authoritative female
`Friendly_Person`	Warm, approachable (default)
`Deep_Voice_Man`	Deep, resonant male
`Calm_Woman`	Soothing, relaxed female
`Casual_Guy`	Informal, conversational male
`Lively_Girl`	Energetic, youthful female

Custom voice cloning: provide a voice sample URL as --voice <url_to_6sec_sample>.

Engine Comparison

Feature	OmniHuman 1.5	Infinite Talk
Cost	$0.12/s (lip-sync)	$0.03/s (480p), $0.06/s (720p)
30s video	~$3.60	~$0.90-$1.80
Max length	30s audio	10 minutes
Quality	Best — full body motion, gestures	Good — solid lip-sync
Resolution	720p/1080p	480p/720p
Best for	Short premium content, ads	Long-form, bulk, prototyping

Full Pipeline Costs (30-second video)

Step	Tool	Cost
Image	nano-banana (Gemini Flash)	~$0.05
Voice	MiniMax Speech 2.6 Turbo	~$0.03
Lip-sync (budget)	Infinite Talk 720p	~$1.80
Lip-sync (premium)	OmniHuman 1.5	~$3.60
Total (budget)		~$1.88
Total (premium)		~$3.68

Cost Tracking

Every generation logs to ~/.avatar-video/costs.json. View summary:

~/.claude/skills/avatar-video/scripts/generate-avatar.sh --costs

API Keys

Service	Key	Source
MiniMax Speech, OmniHuman, Infinite Talk	`FAL_KEY`	`~/.avatar-video/.env`
nano-banana (image gen)	`GEMINI_API_KEY`	`~/.nano-banana/.env`

All fal.ai services use a single key. Get one at https://fal.ai.

Troubleshooting

"FAL_KEY not found" — Run setup: ~/.claude/skills/avatar-video/scripts/setup.sh
"Audio must be under 30s" (OmniHuman) — Use Infinite Talk for longer audio, or trim script
Upload failed — Check FAL_KEY is valid; fal.ai file upload requires auth
Queue timeout — OmniHuman can take 2-5 min; the script polls automatically
nano-banana not found — Install via /nano-banana skill init