FastKoko

FastKoko is a self-hosted text-to-speech server built around the Kokoro-82M model and exposed through a FastAPI backend. It is designed to be easy to deploy via Docker, with separate CPU and GPU images so that users can choose between pure CPU inference and NVIDIA GPU acceleration. The project exposes an OpenAI-compatible speech endpoint, which means existing code that talks to the OpenAI audio API can often be pointed at a Kokoro-FastAPI instance with minimal changes. It supports multiple languages and voicepacks and allows phoneme based generation for more accurate pronunciation and prosody. The server also offers per-word timestamped captions, which makes it useful for creating subtitles or aligning audio with text. A built in web UI, API documentation, and debug endpoints for monitoring system status help users explore voices, test requests, and integrate the service into larger systems.

Features

OpenAI compatible speech API for drop in replacement of cloud TTS
Docker images for both CPU and NVIDIA GPU with multi arch support
Web UI and interactive API docs hosted on the same FastAPI server
Multi language voicepacks with phoneme based synthesis and captions
Voice mixing with weighted combinations and reusable custom voicepacks
Simple quick start scripts and Docker Compose setups for local hosting

Project Samples

Project Activity

See All Activity >

License

Apache License V2.0

Follow FastKoko

FastKoko Web Site

Other Useful Business Software

Propel Software: Product Value Management Platform for Manufacturers

For modern product companies that need to connect product and commercial teams successfully

Propel is a cloud-native Product Value Management platform that unifies PLM, QMS, and PIM in one connected system, giving manufacturers complete visibility and control across the entire product lifecycle. It provides a single source of truth for all product data, streamlines change management, strengthens quality and compliance processes, and accelerates time-to-market by eliminating the silos and manual steps that slow teams down.

Learn More

Rate This Project

User Reviews

Be the first to post a review of FastKoko!

Additional Project Details

Operating Systems

Linux, Mac, Windows

Programming Language

Python

Related Categories

Python Text to Speech Software

Registered

2025-11-28

Similar Business Software

Google AI Studio

Google AI Studio is a unified development platform that helps teams explore, build, and deploy applications using Google’s most advanced AI models, including Gemini 3. It brings text, image, audio, and video models together in one interactive playground. With vibe coding, developers can use...

See Software
Kokoro TTS

Kokoro TTS is an efficient text-to-speech tool with multilingual and customizable voice support. Its 182M parameter architecture delivers high-quality audio, supporting languages like American English, British English, French, Korean, Japanese, and Mandarin. It features lifelike voice options,...

See Software
Murf AI

Murf API is an advanced text-to-speech (TTS) solution that transforms written text into natural, lifelike voiceovers with remarkable accuracy and ease. It empowers developers and businesses with a suite of sophisticated features, including pitch and speed modulation, audio duration adjustments,...

See Software
BeyondWords

BeyondWords is the AI voice platform that brings frictionless audio publishing to writers, newsrooms, and businesses. Every user gets access to 550+ lifelike AI voices across 140+ language locales, and there's the option to commission custom voices. Users can sync their CMS using the API, RSS...

See Software
Fliki

Fliki is a Text to Speech & Text to Video converter that helps you create audio and video content using AI voices in less than a minute. Creating a voice-over isn't an easy task, it's time-consuming, involves days of waiting and is expensive. The same person watches about 30-40 videos in a week...

See Software
EVI 3

Hume AI's EVI 3 is a third-generation speech-language model that streams in user speech and forms natural, expressive speech and language responses. At conversational latency, it produces the same quality of speech as our text-to-speech model, Octave. Simultaneously, it responds with the same...

See Software