skills.rest
wentbackward
wentbackwardCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

Clone Voice Service

Clone voices for TTS and transcribe speech via OpenAI-compatible endpoints.

4|Updated Mar 26, 2026
Search Tags:#fastapi#text-to-speech#whisper#docker#voice cloning#f5-tts#speech-to-text
One-click install
npx skills add https://github.com/wentbackward/clone-voice-service --skill clone-voice-service
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Clone Voice Service
Source: https://github.com/wentbackward/clone-voice-service/tree/main
Command: npx skills add https://github.com/wentbackward/clone-voice-service --skill clone-voice-service

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It provides a simple way to generate cloned-voice text-to-speech and convert uploaded speech back into text, without stitching together separate tools or services.

Core Features & Use Cases

  • Text-to-Speech (TTS) with voice cloning: Generate audio from text using a reference voice profile trained from a 5–15s clip (drop .wav and matching .txt into voices/ for hot-reload).
  • Speech-to-Text (STT) transcription: Transcribe uploaded audio (any common format) into text with configurable language handling and multiple output formats.
  • OpenAI-compatible endpoints: Use /v1/audio/speech and /v1/audio/transcriptions with any OpenAI-style client by pointing baseUrl to the service.
  • Use case: Turn a recorded phrase into consistent synthetic speech for messaging workflows (e.g., generate Telegram/WhatsApp-ready Opus audio) and then transcribe follow-up voice messages for searchable text.

Quick Start

Tell your AI to call POST /tts on your running service to synthesize speech from the text you provide, using a configured voice name.

Frequently Asked Questions about Clone Voice Service

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

FAQPage Schema
How do I clone a voice for text-to-speech using a short audio clip?▼

Voice cloning for text-to-speech is done by dropping a 5–15 second `.wav` clip and matching `.txt` file into the `voices/` directory, allowing the F5-TTS API to hot-reload the profile and synthesize speech from text.

Can I use OpenAI-compatible clients for speech-to-text and text-to-speech APIs?▼

OpenAI-compatible clients work with the `/v1/audio/speech` and `/v1/audio/transcriptions` endpoints by pointing the client `baseUrl` to the service, enabling drop-in compatibility for TTS and STT operations.

How do I transcribe uploaded audio files into text using Whisper?▼

Speech-to-text transcription is handled via the `/stt` endpoint using Whisper, which automatically converts uploaded audio from any common format into 16kHz mono WAV before running inference to output text.

Does FastAPI support serving binary audio outputs for messaging workflows?▼

FastAPI supports serving binary audio outputs directly from the TTS API, defaulting to OGG Opus format to provide Telegram and WhatsApp-ready voice messages.

What audio format do I need for speech-to-text transcription inference?▼

Speech-to-text transcription requires audio to be converted to 16kHz mono WAV format before Whisper inference, a conversion the API handles automatically from any common uploaded format.

Do I need Docker to deploy a self-contained voice cloning and transcription API?▼

Docker is listed as a deployment keyword for this self-contained API, which combines F5-TTS voice cloning and Whisper transcription into a single FastAPI service without stitching separate tools.

Related Skills

Explore More Software Engineering Skills→
GallonShihGallonShih
3

Docker Compose Development

Analyze docker-compose.yml to optimize mounted-code development workflows.

Community
Advanced
Prathviraj-jadhavPrathviraj-jadhav

ASR

Transcribe spoken audio from files or streams into structured text.

Community
Intermediate
jayll1303jayll1303
17

openai-audio-api

Build OpenAI-compatible TTS endpoints with streaming and batch inference.

Community
Advanced
mxzinkemxzinke
2

stt

Transcribe audio recordings to text via a Whisper-compatible STT backend.

Community
Intermediate