CoolFace
Datasetpublic

Muckylixx/german-sohee-synthetic-tts-24k

Flevi Restiti Vici Bis uns der Weg nur noch nach vorn blieb This repository contains a synthetic German audiobook and text-to-speech training dataset based on the original novel: Flevi Restiti ViciBis uns der Weg nur noch nach vorn blieb The novel was written in German by Maurice Hartmann, who is the author and copyright holder. Work in progress This dataset is a work in progress. Future revisions may include corrected transcripts, regenerated audio… See the full description on the dataset page: https://huggingface.co/datasets/Muckylixx/german-sohee-synthetic-tts-24k.

sourceHugging Faceotherupdated 3mo agoView on Hugging Face
0likes389downloads
Dataset Card

Flevi Restiti Vici

Bis uns der Weg nur noch nach vorn blieb

This repository contains a synthetic German audiobook and text-to-speech training dataset based on the original novel:

Flevi Restiti Vici Bis uns der Weg nur noch nach vorn blieb

The novel was written in German by Maurice Hartmann, who is the author and copyright holder.

Work in progress

This dataset is a work in progress. Future revisions may include corrected transcripts, regenerated audio clips, pronunciation corrections, improved pauses and emphasis, metadata corrections, structural improvements and additional documentation.

File counts, duration and individual clips may change in later revisions.

Dataset summary

  • —Language: German
  • —Chapters: 58
  • —Number of clips: 7,709
  • —Total duration: approximately 19.46 hours
  • —Sample rate: 24,000 Hz
  • —Audio format: WAV
  • —Speaker preset: Sohee
  • —Source text: original German novel
  • —Audio source: synthetically generated

Generation

The audio was generated with Qwen3-TTS-12Hz-0.6B-CustomVoice using the built-in Sohee speaker preset.

This repository is an independently created dataset and audiobook project. It is not an official release by the Qwen team.

Copyright and permitted private use

The complete transcript originates from the original German novel:

Flevi Restiti Vici – Bis uns der Weg nur noch nach vorn blieb

Maurice Hartmann is the author and copyright holder.

Private and non-commercial use is permitted, including private reading, private listening, private TTS model training, private speech-synthesis experiments, private conversion into other audio formats, private translation into other languages, private generation of translated audio or audiobooks, and personal research and technical experimentation.

Privately created translations may be read, generated and listened to for personal purposes.

Public publication, commercial distribution, resale or presentation as an official translated edition requires explicit permission from Maurice Hartmann.

Other language editions

Anyone interested in creating or publishing an official version of the novel in another language is welcome to contact Maurice Hartmann through this repository.

Possible cooperation may include translations, translated audiobooks, public distribution, commercial use and official editions in other languages.

File naming

The files are numbered sequentially.

Example:

text
00001_C01.wav
00001_C01.txt

Meaning:

  • —00001 is the sequential clip number
  • —C01 means chapter 1
  • —C02 means chapter 2
  • —C58 means chapter 58
  • —.wav contains the synthetic speech
  • —.txt contains the corresponding transcript

Sorting the files by filename restores the original order of the novel.

Directory structure

text
dataset/
  C01/
    00001_C01.wav
    00001_C01.txt
    ...
  C02/
    ...
  ...
  C58/
    07709_C58.wav
    07709_C58.txt

manifest.jsonl
README.md
LICENSE.md

Each WAV file has a transcript file with the same base filename.

Audiobook reconstruction

Playing or concatenating the WAV files in sequential filename order reconstructs the synthetic German audiobook.

The reconstructed audiobook may be created, converted and listened to privately.

Public audiobook publication, commercial release or redistribution requires permission from Maurice Hartmann.

Limitations

The audio is synthetically generated and may contain pronunciation errors, unusual pauses, inconsistent emphasis, generation artifacts, occasional transcript/audio mismatches and varying speaking rhythm between clips.

Author and copyright holder

Maurice Hartmann

Published on Hugging Face under:

Muckylixx / Mucky Works