Skip to content
TakoAPI
Back to skills

clonev

by instant-picture

Speech & Transcription
1.7kDownloads reported by clawskills.sh
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Clone any voice and generate speech using Coqui XTTS v2.

Description

What This Skill Does

Clones any voice from a short WAV audio sample and generates new speech in that voice using Coqui XTTS v2. Accepts text, a WAV sample (6-30 seconds), and a language code, then outputs an OGG audio file. Supports 16 languages including English, Czech, German, French, Spanish, Japanese, and Korean.

A single script call handles model loading, Docker orchestration, and file output automatically, with no API keys or external services required.

When to Use It

  • Generating personalized voice messages without re-recording each time
  • Making a character or persona speak custom text in their voice
  • Producing multilingual audio from a single voice sample
  • Creating voiced task notifications or reminders in a familiar voice
  • Turning written scripts into audio with a specific person's voice

Example Workflow

INPUT

User asks: Say 'Project delivered on time' in my voice

AGENT
  1. 1

    Ask user for the WAV sample path, confirm language code defaults to 'en'

  2. 2

    Run scripts/clonev.sh with the text, sample path, and language code

  3. 3

    Wait 20-40 seconds for XTTS v2 processing to complete

  4. 4

    Read the returned OGG file path from the script output

OUTPUT

OGG audio file saved to /mnt/c/TEMP/Docker-containers/coqui-tts/output/clonev_output.ogg

Requirements

  • Docker installed on the system

Installation

clawhub install instant-picture/clonev

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links