Books deserve a voice.

Multi-voice synthesis

Assign unique voices to each character. Narration, dialogue, and inner monologue each get their own voice.

Dialogue detection

Automatically identifies quoted dialogue vs narration and attributes speech to the correct speaker.

Emotion inference

Rule-based and lexicon-powered emotion labeling gives each line the right tone and delivery.

Voice suggestions

Explainable, ranked voice recommendations for every speaker based on character traits and context.

Review before render

Human-editable review format lets you correct attributions and emotions before committing to audio.

Persistent cache

Resume failed renders without re-synthesizing completed chapters. Pick up right where you left off.

Install

# Clone and install
git clone https://github.com/mcp-tool-shop-org/audiobooker
cd audiobooker
pip install -e .

# Required: voice-soundboard for TTS
pip install -e ../voice-soundboard

# Required: FFmpeg for audio assembly
# Windows: winget install ffmpeg
# Mac: brew install ffmpeg

Quick workflow

# Create project from EPUB
audiobooker new mybook.epub

# Get voice suggestions & auto-apply
audiobooker cast-suggest
audiobooker cast-apply --auto

# Compile, review, render
audiobooker compile
audiobooker review-export
audiobooker render

From EPUB

from audiobooker import AudiobookProject

project = AudiobookProject.from_epub("mybook.epub")
project.cast("narrator", "bm_george", emotion="calm")
project.cast("Alice", "af_bella", emotion="warm")
project.compile()
project.render("mybook.m4b")

From text

from audiobooker import AudiobookProject

project = AudiobookProject.from_string(
    "Chapter 1\n\nHello world.",
    title="My Book"
)
project.compile()
project.render("mybook.m4b")