Search is not available for this dataset
audio
audioduration (s) 28.6
44.3
|
|---|
YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/datasets-cards)
Voxly Voice Dataset
High-quality voice samples for XTTS voice cloning.
Dataset Information
- Total Voices: 88
- Sample Duration: 30+ seconds per voice
- Format: WAV (16-bit PCM, 22050 Hz)
- Languages: 20+ languages including English, Spanish, French, German, Arabic, Chinese, Japanese, Korean, and more
- Accents: Authentic regional accents (American, British, Australian, Canadian, Irish, South African, etc.)
Voice Categories
English Variants
- American English (12 voices)
- British English (5 voices)
- Australian English (2 voices)
- Canadian English (2 voices)
- Irish English (2 voices)
- Indian English (2 voices)
- South African English (2 voices)
European Languages
- Spanish (6 voices - Spain, Mexico, Argentina)
- French (4 voices - France, Canada)
- German (4 voices - Germany, Austria)
- Italian (2 voices)
- Portuguese (4 voices - Brazil, Portugal)
- Russian (2 voices)
- Polish (2 voices)
- Dutch (2 voices)
- Swedish (2 voices)
- Norwegian (2 voices)
- Danish (2 voices)
- Finnish (2 voices)
- Greek (2 voices)
- Czech (2 voices)
- Turkish (2 voices)
Asian Languages
- Chinese (4 voices)
- Japanese (2 voices)
- Korean (2 voices)
- Hindi (2 voices)
- Thai (2 voices)
- Vietnamese (2 voices)
Middle Eastern Languages
- Arabic (3 voices - Saudi, Egyptian)
Usage
These voice samples are optimized for XTTS (Coqui TTS) voice cloning to produce natural, human-sounding speech synthesis.
Voice Naming Convention
Files are named using the pattern: {language}_{region}_{voicename}neural.wav
Example: en_us_arianeural.wav = English (US), Aria voice
- Downloads last month
- 10
Size of downloaded dataset files:
18.4 MB
Size of the auto-converted Parquet files:
16.6 MB
Number of rows:
88