Slayprincess: Voice-Acted Audio Lines from a Video Game

TeraTTS provides a dataset of 9,394 high-quality audio clips paired with transcript text, extracted from the video game Slay the Princess. The collection totals approximately 13 hours of audio across three primary speakers. The dataset was last updated on Hugging Face in May 2026.

Use Cases

• Training text-to-speech models based on the high-quality, character-specific voice recordings.
• Fine-tuning speech recognition systems on the paired audio and transcript data.
• Developing voice conversion or cloning systems using the distinct speaker profiles (narrator, princess, hero).
• Creating audio datasets for game development or interactive media based on the professionally acted lines.

Strengths

• Contains 9,394 individual audio clips, providing a substantial volume of data.
• Audio is high-quality FLAC format at 44.1 kHz, suitable for detailed analysis.
• Total duration is approximately 13 hours, offering extensive material for training.
• Data is structured with clear speaker labels (narrator, princess, hero) and clip counts for each.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Access is gated behind a non-commercial research agreement, which may restrict use.

Provenance

Source

TeraTTS on Hugging Face.

Collection Method

Extracted from the video game Slay the Princess.

Freshness

Last updated 2026-05-20 17:49:09; freshness should be verified.