Hokuspokus Qwen3 Tts Hybrid: German TTS Corpus with Human and Synthetic Audio

A professional-grade German text-to-speech training corpus created by author semidark and last updated on 2026-04-13. It combines high-quality human narration from the HUI Audio Corpus and LibriVox with synthetic augmentation to provide a legally safe alternative for training models like kokoro. The dataset is described as a work in progress.

Use Cases

• Train German text-to-speech models based on the described high-quality human narration.
• Fine-tune speech synthesis models based on the state-of-the-art synthetic augmentation mentioned.
• Create legally safe training data for voice AI based on the combined human and synthetic sources.

Strengths

• Combines high-quality human narration from professional narrator 'Hokuspokus'.
• Includes state-of-the-art synthetic augmentation for data expansion.
• Designed to be a legally safe alternative for model training.

Limitations

• Row count, file formats, and column-level documentation are unknown.
• Dataset is explicitly marked as a work in progress.
• License information is unknown, which may restrict usage.

Provenance

Source

HUI Audio Corpus and LibriVox, with voice seed from professional narrator 'Hokuspokus'.

Collection Method

Combination of human narration and synthetic augmentation.

Time Range

null

Freshness

Last updated 2026-04-13 06:19:27.

Geography

null