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

ai-voice-cloning

Generates natural AI speech from text using inference.sh CLI TTS models.

Updated Jul 28, 2026
Search Tags:#text-to-speech#elevenlabs#tts#voiceover#narration#voice generation#kokoro
One-click install
npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill ai-voice-cloning-aadi-110i
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-voice-cloning
Source: https://github.com/Aadi-110i/PEP-PROJECT/tree/main/skills/ai-voice-cloning
Command: npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill ai-voice-cloning-aadi-110i

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Creating voiceovers, audiobook narration, or podcast audio traditionally requires recording equipment, voice talent, and editing time. This Skill generates natural-sounding speech from text via the inference.sh CLI, covering multiple TTS models and voice styles. ## Core Features & Use Cases - Multiple TTS Models: Access ElevenLabs (22+ voices, 32 languages), Kokoro TTS, DIA, Chatterbox, Higgs, and VibeVoice through a unified CLI interface. - Voice Library & Control: Choose from American and British English voices with adjustable speed (0.8-1.2) and punctuation-based pacing control. - Long-Form & Multi-Voice Workflows: Chunk long texts, merge audio segments, create multi-speaker conversations, and combine voiceovers with video or talking-head avatars. - Use Case: A content creator needs narration for a 10-minute tutorial video. They generate the voiceover with a professional Kokoro voice, then merge it with their video using the media-merger app. ## Quick Start Use the ai-voice-cloning skill to generate a warm female voiceover reading my intro paragraph with the Kokoro af_sarah voice.

Frequently Asked Questions about ai-voice-cloning

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

FAQPage Schema
How do I generate AI text-to-speech audio from the command line?▼

Install the inference.sh belt CLI, run belt login, then execute belt app run with a TTS app like infsh/kokoro-tts, passing a JSON input with your prompt text and voice ID. The command returns a URL to the generated audio file.

What TTS model should I use for audiobook narration?▼

Kokoro TTS with voices like bf_emma at speed 0.8-0.9 works well for audiobooks, while Higgs targets professional narration. For premium quality across 32 languages, ElevenLabs TTS offers 22+ voices.

Can Kokoro TTS handle long-form text over 5000 characters?▼

Long content should be split into chunks under 5000 characters and processed separately. Each chunk generates its own audio, which you then merge using the infsh/media-merger app with a crossfade setting.

How do I create a multi-speaker AI conversation?▼

Generate each speaker's lines separately using different voice IDs, such as am_michael and af_sarah. Then merge the resulting audio URLs with infsh/media-merger using a crossfade of around 300ms for natural transitions.

Does AI voice generation work with video content?▼

Yes, generate the voiceover audio first, then combine it with video using infsh/media-merger by passing video_url and audio_url. You can also create talking-head videos by pairing speech with bytedance/omnihuman-1-5 and a portrait image.

Related Skills

Explore More Content & Communication Skills→
tomada1114tomada1114
4

video-script-writing

Generate structured YouTube scripts from Markdown articles, notes, or outlines.

Community
Basic
inference-shinference-sh
723

ai-podcast

Generates multi-speaker talking head podcast videos via portrait, TTS, avatar, and merge pipeline.

Official
Advanced
47thtechcorner47thtechcorner
21

elevenlabs

Generate voiceovers, sound effects, and music via the ElevenLabs Python client.

Community
Advanced
shalevaminshalevamin

sherpa-onnx-tts

Generates speech audio from text using the offline sherpa-onnx TTS engine.

Community
Intermediate