---
license: cc0-1.0
size_categories:
- n<1K
tags:
- embedding
- metadata
pretty_name: Voice Converted Arabic Common Voice 12.0 Speakers
dataset_info:
  features:
  - name: embedding
    sequence: float32
  - name: male
    dtype: bool
  - name: reference
    dtype: string
  - name: other
    sequence: int32
  - name: mostly_corrupted
    dtype: bool
  splits:
  - name: train
    num_bytes: 27386
    num_examples: 26
  download_size: 27724
  dataset_size: 27386
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

These are the speakers' embeddings and information for the [Voice Converted Arabic Common Voice 12.0](https://hf.co/datasets/xmodar/commonvoice-12.0-arabic-voice-converted) dataset.