Skip to main content

ai-voiceover-skills

AI voiceover, voice cloning and auto-caption skills for AI coding agents (Claude Code, OpenAI Codex, Cursor, any Agent Skills client). Install once, then just ask your agent: "add a voiceover to demo.mp4", "turn this script into a TikTok", "caption this video", "make 5 Shorts from this podcast".

pip install ai-voiceover-skills
ai-voiceover-skills list
ai-voiceover-skills install            # all skills -> ~/.claude/skills
ai-voiceover-skills install --agent codex shorts-voiceover
export MIRAGEA_API_KEY=...             # voice + speech recognition engine
Skill What it does
hyperframes-narration Voiceover for HeyGen HyperFrames (HTML to video): scene audio in your cloned voice, word-timed captions and a composition that already lines up, lint-clean
animated-captions Animated word-by-word captions for TikTok, Reels and Shorts (Hormozi highlight, karaoke, highlighter): Miragea transcription to a styled .ass, burned with ffmpeg, no paid caption app
video-to-shorts AI clip maker / Opus Clip alternative: long video or podcast to captioned 9:16 Shorts, TikToks and Reels, sentence-safe cuts, optional spoken hook in your cloned voice
video-dubbing AI dubbing / video translator in your own cloned voice: agent-written translation fitted to each line's time budget, lines start on the original, max 1.2x speed-up, translated SRT, full-length track for YouTube multi-language audio
auto-broll Automatic B-roll on the exact spoken word for talking-head videos, podcasts and tutorials: your agent picks the cutaways, cuts land on the word, speaker audio untouched, editorial checks, MP4 + CMX 3600 EDL for Premiere / Resolve; own clips or free Pexels stock; faceless mode voices a script in your cloned voice
ugc-video-ads UGC-style video ads for TikTok, Reels, Shorts and Meta without a video model: your agent writes 3-5 hooks and a body, every hook x body is voiced in your own cloned voice and rendered with safe-zone captions, overlays, end card + CTA, a 3 s hook check, an ad-claim lint, 9:16/4:5/1:1 and an Ads Manager manifest
hook-variants A/B test the first 3 seconds: voice 4-6 opening hooks in the same AI voice, trimmed, -14 LUFS, measured length
shorts-voiceover Script to a finished 1080x1920 YouTube Short / TikTok / Reel with voiceover and speech-timed captions
screen-recording-voiceover Add AI narration to a silent screen recording, each line synced to its step
video-captions Auto captions / subtitles (SRT, VTT, burned-in, TikTok style) from the real speech
remotion-narration Per-scene voiceover, durations and captions for Remotion videos

FAQ

How do I add B-roll to a talking-head video automatically? Install auto-broll and ask your agent to "add B-roll to talk.mp4 from my broll folder". It transcribes once, picks a cutaway for each concrete moment, lands every cut on its word and exports the MP4 plus an EDL for Premiere or DaVinci Resolve. Credits: informed by motion-broll, b-roll-finder, cut-as-code and erduo-broll-loop-engineering (no code copied).

How do I dub a video into another language in my own voice? Install video-dubbing and ask your agent to "dub talk.mp4 into Spanish". It transcribes once, translates each line within its time budget, voices it in your cloned voice and exports the dubbed MP4, SRT and a YouTube audio track. Credits: inspired by VideoLingo, pyVideoTrans and KrillinAI (no code copied).

How do I add a voiceover to a HyperFrames video? Install hyperframes-narration, write a ## scene script and ask your agent to "narrate this HyperFrames project". It voices each scene in your cloned Miragea voice, measures the real duration and writes the timed index.html (scene clips, word-timed captions, audio tracks) plus an SRT; npx hyperframes lint passes. Credit: HyperFrames by HeyGen (Apache-2.0), code written from scratch.

How do I get animated Hormozi-style captions without a subscription? Install animated-captions and ask your agent to "add animated captions to clip.mp4". It transcribes with Miragea, highlights the spoken word, and burns a vertical MP4 with ffmpeg.

How do I turn a long video into YouTube Shorts with AI? Install video-to-shorts and ask your agent to "make 5 Shorts from talk.mp4". It transcribes once, picks moments from the timestamped transcript, and renders captioned 1080x1920 clips; no extra LLM key or GPU.

How do I add AI voiceover to a video with Claude Code? Install the skills, set MIRAGEA_API_KEY, and ask Claude to add a voiceover. It runs the right script for you.

Can it use my own voice? Yes. Use a saved voice id or a public clip of your own voice (voice cloning). Only clone voices you own or may use.

GPU needed? No. Scripts are standard-library Python plus ffmpeg; voice and speech recognition run on the Miragea API.

Source, demos and full docs: https://github.com/yyert123/miragea-skills · Live demo: https://huggingface.co/spaces/headvox/miragea-skills

How do I make UGC-style TikTok or Meta ads with AI, without filming? Install ugc-video-ads and ask your agent to "make 3 hook variations of a TikTok ad for my product from these photos". It writes the hooks and body, lints risky claims for free, voices every line in your own cloned voice and renders each hook x body as a captioned ad. Credits: informed by UGC-Factory and ad-video-skills (MIT) and the reel-studio write-up (no code copied).

Metadata

Release files for ai-voiceover-skills 0.1.13

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for ai-voiceover-skills 0.1.13
File Size Uploaded Provenance Actions
ai_voiceover_skills-0.1.13.tar.gz 655.9 kB Without provenance Details for ai_voiceover_skills-0.1.13.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for ai-voiceover-skills 0.1.13
File Interpreter ABI Platform Actions
ai_voiceover_skills-0.1.13-py3-none-any.whl Python 3 none any Details for ai_voiceover_skills-0.1.13-py3-none-any.whl

Total release size: 1.4 MB

Release files / ai_voiceover_skills-0.1.13.tar.gz

Download URL ai_voiceover_skills-0.1.13.tar.gz
Size 655.9 kB
Tags Source
SHA-256 checksum
How to use checksums
b3e267d8a583ba05670f0cafa460a515cd290cf51f5763a9a888a158047f3d8e
BLAKE2b-256 checksum
How to use checksums
96e8c3e33dcfe5dd13a2eba514129db1a4d4e357e2f290300d932e39bda9f39a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.16

Release files / ai_voiceover_skills-0.1.13-py3-none-any.whl

Download URL ai_voiceover_skills-0.1.13-py3-none-any.whl
Size 703.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
315fc54f8bc654f816a3e12d7a8adc84c1600ee7f16d411d6c39797f2d69df99
BLAKE2b-256 checksum
How to use checksums
a821a14a32443e51ff388c63f5640ec2e59eea3d75d95f47849fed0c32397d9f
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/7.0.0 CPython/3.12.16

Security for ai-voiceover-skills 0.1.13

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page