SK-404

Doc no. SK-404DatasheetRev 2026.11Sheet 2.11 of 4

A5

Odysseus Speech Tools

Glue and tools for a local, OpenAI-compatible speech engine in Docker, built for Odysseus: Kokoro-82M speech, Faster-Whisper transcription and document-to-audio scripts.

Active

01

Block diagram

DocumentsPDF · MD · EPUB narrator.py→ MP3 pdf2epub_tts.py→ .m4b + EPUB Odysseusvoice in and out Speaches engineDocker · OpenAI API Kokoro-82Mtext to speech · ONNX Faster-Whisperspeech to text local · CPU only
Fig. 1: One speech engine, three front doors
02

Characteristics

Text-to-speechKokoro-82M (ONNX): neural voices with natural prosody
Speech-to-textFaster-Whisper (Systran/faster-whisper-small by default), local and memory-efficient
EngineThe open-source Speaches server in a Docker container: an OpenAI-compatible API, CPU only
Narratornarrator.py reads PDF, Markdown, EPUB and text aloud to MP3
Audiobookspdf2epub_tts.py strips headers, footers, page numbers and citations, then builds a chaptered .m4b and a clean EPUB
Verificationverify_all.py and test_audio_engine.py check the engine end to end
  • Docker
  • Kokoro-82M
  • Faster-Whisper
  • Python
03

Overview & build notes

Connects Odysseus to a local Speaches Docker container serving OpenAI-compatible /v1/audio/speech and /v1/audio/transcriptions endpoints (http://localhost:8000/v1) powered by Kokoro-82M (ONNX) and Faster-Whisper.

Alongside real-time agent voice replies, the repository includes narrator.py for quick PDF/Markdown/EPUB-to-MP3 readout and pdf2epub_tts.py (which evolved into Kokoro Audiobook Studio) for chaptered .m4b conversion.

Both verify_all.py and test_audio_engine.py validate live model aliases, TTS/STT round-trips and container health end to end.

04

Files & artifacts

KindFileDetailsSize
Scriptnarrator.pyDocument-to-MP3 CLI reader supporting PDF, Markdown, EPUB and plain textGitHub
Scriptpdf2epub_tts.pyLayout cleaner and chaptered .m4b / EPUB generator on GitHubGitHub
Testsverify_all.pyEnd-to-end TTS/STT engine verification suite on GitHubGitHub
DocsPDF2EPUB_TTS_README.mdUsage guide for document cleanup and multi-chapter audiobook buildsGitHub

A1Kokoro Audiobook Studio