Pre-Release · Windows Desktop Suite · 100% Free & Offline

Offline Transcription & Zero-Shot Voice Cloning

Transcribe audio to text and generate realistic voice clones locally on Windows. Powered by Whisper, Chatterbox, Kokoro, LuxTTS, TADA, and Qwen — no cloud required.

Pre-Release Status: CincoScribe is in active pre-release development. Custom Voice Cloning & CUDA GPU acceleration integration are in active progress.
Download Free Explore Engines
CincoScribe Desktop Workstation
C
Preferences & Workstation
Manage local sidecar server, GPU acceleration, and system settings.
Local sidecar backend: http://127.0.0.1:5555 · Status: ONLINE
Server Connection: ONLINE · Port 5555
Custom Voice: Active Dev | CUDA: In Progress
0ms Cloud Latency (100% Local Inference) Zero Data Uploads or Account Tracking CUDA & CPU Auto-Dispatch Support Zero-Shot Voice Cloning from 5s Reference Audio Native Expression Tags [laugh] [cough] [chuckle] 24+ Language ASR & Real-time Audio Streaming 0ms Cloud Latency (100% Local Inference) Zero Data Uploads or Account Tracking CUDA & CPU Auto-Dispatch Support Zero-Shot Voice Cloning from 5s Reference Audio Native Expression Tags [laugh] [cough] [chuckle] 24+ Language ASR & Real-time Audio Streaming

How It Works

Zero cloud uploads. Zero account setup. Full local processing.

1

Install Desktop App

Download the free Windows executable. Includes local sidecar engine wrappers.

2

Select Engine

Choose Whisper for ASR or Chatterbox / Kokoro for zero-shot voice cloning.

3

Process Locally

Generate speech or transcribe audio instantly on CPU/GPU without internet.

Supported Local Model Engines

Declarative backend architecture powering offline speech and voice capabilities.

ASR · Speech To Text

Faster-Whisper (ONNX / CTranslate2)

Tiny, Base, Small, Medium, Large-v3, and Turbo variants for 24+ language transcription and real-time streaming.

TTS · Multilingual Voice Cloning

Chatterbox Multilingual

Zero-shot voice cloning across 23 languages using 5-second reference audio prompts.

TTS · Expression Tags

Chatterbox Turbo

Fast English voice cloning with native support for paralinguistic tags like [laugh], [cough], and [chuckle].

TTS · Lightweight 82M

Kokoro 82M

High-fidelity, ultra-lightweight speech synthesis with minimal memory footprint.

TTS · Fast CPU

LuxTTS

CPU-friendly voice generation designed for fast execution without dedicated GPU requirements.

TTS & LLM · Multimodal

TADA & Qwen Engine Backends

HumeAI TADA (1B & 3B) text-acoustic dual alignment and Qwen2 Audio custom voice backends.

Pricing & Support

CincoScribe is 100% free and open-source under the MIT License.

Desktop Application
Free
MIT License · Fully Offline
  • Unlimited offline transcriptions
  • Zero-shot voice cloning
  • No license keys or paywalls
  • Full data privacy and control
Download Free
Support Project
Donate
Support open source development

If CincoScribe helps your workflow, consider supporting ongoing model and feature development on Ko-fi.

Buy Me a Coffee at ko-fi.com

Frequently Asked Questions

No. All transcription and text-to-speech synthesis run locally on your device sidecar without sending data online.
CincoScribe supports Chatterbox Multilingual, Chatterbox Turbo, Kokoro 82M, LuxTTS, TADA, and Qwen CustomVoice.
Portfolio Home