The Cebuano Speech Dataset provides 108 hours of audio data across 807 files in MP3 and WAV formats. It was created by Speech-data and includes balanced voice data with 49% female and 51% male speakers aged 18 to 50+ years.
Use Cases
- Train speech recognition models based on the 108 hours of diverse audio.
- Develop text-to-speech systems based on the balanced gender distribution of speakers.
- Create voice cloning applications based on the dataset's structured audio files.
- Benchmark audio processing algorithms based on the provided MP3 and WAV formats.
Strengths
- Contains 108 hours of audio data, providing substantial material for training.
- Includes 807 individual audio files, offering a structured collection.
- Features a balanced gender distribution with 49% female and 51% male speakers.
- Covers a broad age range from 18 to 50+ years, suggesting demographic diversity.
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.
Provenance
- Source
- Speech-data
- Freshness
- Last updated 2026-03-30 12:41:12; freshness should be verified.