TTS-Story

Free plan APILinuxMacSelf-hostedWebWindows

What TTS-Story is

TTS-Story is a free, web-based text-to-speech application for producing narrated stories and audiobooks. It offers twenty generation choices spanning local CPU, local GPU, self-hosted, and cloud speech engines. Speaker tags support multi-voice narration, and compatible engines offer reference voice cloning and voice-design workflows. Optional manuscript preparation connects to Gemini, Atlas Cloud, OpenRouter, LM Studio, or Ollama. For longer works, the app detects chapters and sections, allowing separate chapter exports or one combined output. Its job queue supports progress tracking, pausing, resuming, cancellation, retries, and recovery checkpoints. The Audio Library supports chunk review, regeneration, rebuilding, and MP3 or M4B exports; other listed formats are WAV and OGG. Per-speaker pitch and speed controls, pause markers, silence controls, and ACX-oriented processing are also included. Local engines keep manuscript text and synthesis on the computer after required downloads, while cloud providers receive text for synthesis and cloud cloning services may receive reference audio. Settings and API keys are stored locally in config.json as plain text, which users are advised not to share. Some heavier local engines need substantial GPU memory or may be impractical on CPU.

Who it suits

TTS-Story may suit creators making multi-voice stories or audiobooks who want local, self-hosted, or cloud engine choices. Users should consider the project’s note about plain-text credential storage and the demands of some local engines.

What it does well

Twenty generation choices across four deployment types
Supports multi-speaker narration and compatible voice cloning
Detects chapters for separate or combined exports
Queue supports pause, resume, retries, and recovery
Exports MP3, WAV, OGG, and M4B

What to know first

Settings and API keys are stored as plain text
Some local engines need substantial GPU memory
Cloud providers receive submitted synthesis text

Limits

The application is free, but hosted engines can require provider accounts, credits, subscriptions, or usage billing. Some heavier local engines require substantial GPU memory and may be very slow or impractical on CPU; commercial hosted Breeze output requires an active paid subscription at generation time.

Plans

Plan	Billed and limits	Price
Free	Application freely available · hosted engines may require provider accounts, credits, subscriptions, or usage billing	Free

As github.com lists them, read 13 Oct 2026. The maker's own page has the final word.

Getting started

Open the TTS-Story GitHub website
Install it using the Windows, Linux, macOS, or Pinokio instructions
Choose and install local engines individually, or connect a hosted provider
Prepare a manuscript and optionally connect Gemini, Atlas Cloud, OpenRouter, LM Studio, or Ollama
Use speaker tags, detect chapters, and queue narration jobs
Review chunks in the Audio Library and export the finished audio

Verdict

TTS-Story combines audiobook-oriented editing and export tools with a broad selection of speech generation approaches. Users choosing cloud services should account for text sharing, and users choosing local engines should check their hardware demands.

Questions

Is TTS-Story free?

Yes. The Free plan is listed as free, and the application is freely available. Hosted engines may still require provider accounts, credits, subscriptions, or usage billing.

Which platforms can run TTS-Story?

Listed platforms are API, Linux, macOS, self-hosted, web, and Windows. Installation instructions also include a Pinokio option.

Can it clone voices?

Yes. Compatible engines support reference voice cloning and voice-design workflows. Cloud cloning services may receive reference audio.

What audio formats can it export?

TTS-Story lists MP3, WAV, OGG, and M4B exports. It can export separate chapters or a combined full-story output.

Where is manuscript data processed?

Local engines keep manuscript text and synthesis on the computer after required downloads. Cloud providers receive the text they synthesize.

What license covers the application?

The application code is licensed under Apache 2.0. Third-party code, model weights, voice samples, and hosted services retain their own terms.

Facts from TTS-Story's own pages

Audiobook tools: It detects chapters and sections and supports separate chapter exports or a combined full-story output. · github.com, 12 Oct 2026
Cloud providers: Listed hosted speech connections include Replicate, Microsoft Azure Speech, Microsoft Edge TTS, ElevenLabs, Breeze API, and OpenAI-compatible TTS. · github.com, 12 Oct 2026
Commercial limits: Commercial hosted Breeze outputs require an active paid subscription at generation time, and local Breeze model restrictions also apply. · github.com, 13 Oct 2026
Credential storage: Settings and API keys are stored in local config.json as plain text, and the maker advises users not to share or commit that file. · github.com, 13 Oct 2026
Engine choices: It offers twenty normal generation choices across local CPU, local GPU, self-hosted, and cloud generation. · github.com, 12 Oct 2026
Engines: It offers twenty generation choices across local CPU, local GPU, self-hosted, and cloud speech generation. · github.com, 13 Oct 2026
Export formats: MP3,WAV,OGG,M4B · github.com, 20 Sept 2026
Export processing: It includes per-speaker pitch and speed controls, pause markers, silence controls, and ACX-oriented processing. · github.com, 13 Oct 2026
Free plan: Yes · github.com, 20 Sept 2026
Installation: The project provides installation instructions for Windows, Linux, and macOS, plus a Pinokio installation option. · github.com, 12 Oct 2026
License: The application code is licensed under Apache 2.0, while third-party models, voice samples, and hosted services have separate terms and may not grant commercial output rights. · github.com, 13 Oct 2026
Limits: The README notes that some heavier local engines need substantial GPU memory and can be very slow or impractical on CPU. · github.com, 12 Oct 2026
Local engines: Local engines can be installed individually, each in its own virtual environment and model/cache folder. · github.com, 12 Oct 2026
Narration: Speaker tags support multi-speaker narration, and voice options include built-in voices, custom blends, reference cloning, and voice design workflows. · github.com, 13 Oct 2026
Platforms: The maker documents installation on Windows, Linux, and macOS, plus installation through Pinokio. · github.com, 13 Oct 2026
Privacy: Local engines keep manuscript text and synthesis on the computer after required downloads, while cloud TTS and LLM providers receive the text sent to them. · github.com, 13 Oct 2026
Production features: The app detects chapters, queues jobs with pause and resume, and provides chunk review, speaker regeneration, MP3 downloads, and M4B packaging. · github.com, 13 Oct 2026
Purpose: TTS-Story is a web-based multi-voice text-to-speech application for creating narrated stories and audiobooks. · github.com, 13 Oct 2026
Queue and library: The job queue supports progress, pause and resume, cancellation, retries, and recovery checkpoints, while the Audio Library supports chunk review, regeneration, rebuilding, and MP3/M4B exports. · github.com, 12 Oct 2026
Remote administration: Engine installation, removal, and backend restart are localhost-only unless authenticated remote administration is explicitly enabled. · github.com, 13 Oct 2026
Self-hosted connection: TTS-Story can connect to an existing self-hosted LocalAI server and discover compatible TTS models and server voices. · github.com, 13 Oct 2026
Support: The maker directs users to open GitHub issues for bugs, feature requests, or questions and provides a searchable in-app Help tab. · github.com, 13 Oct 2026
Text preparation: Optional manuscript preparation connects to Gemini, Atlas Cloud, OpenRouter, LM Studio, or Ollama. · github.com, 13 Oct 2026
Voice cloning: Yes · github.com, 20 Sept 2026

Where TTS-Story stands on Freedom251

Text-to-speech tools#37 of 232
Text-to-Speech Software#92 of 223