metadata
datasets:
- elevenlabsSpeechTest
description: Speech dataset generated using ElevenLabs API
language:
- en
features:
- file_name: path to audio file
- transcription: the spoken text in the audio file
- speaker: speaker identifier
- audio: audio file in mp3 format
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
dataset_info:
features:
- name: file_name
dtype: string
- name: transcription
dtype: string
- name: speaker
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 3061235
num_examples: 800
- name: test
num_bytes: 747574
num_examples: 200
download_size: 3673462
dataset_size: 3808809
ElevenLabs Speech Dataset
This dataset contains speech data generated using the ElevenLabs API. It includes phrases in various variants, processed with different stability settings and recorded by multiple speakers.
Data Description:
- Transcriptions: Each transcription corresponds to a phrase that was spoken by one of 10 different speakers.
- Speakers: 10 different speakers were used.
- Stability Levels: 5 stability levels were applied to each transcription.
- Variants: The dataset includes multiple variants of each phrase.
Usage:
You can use this dataset for training speech recognition models like wav2vec2 from Hugging Face.
Train / Test Split:
The dataset is split into 80% training and 20% testing.
Audio Files:
Each audio file is named using the format:
{part_name}_{column}_{index}_{speakID}_style{stableVal}.mp3
where:
part_name: The name of the part from the CSV file.column: The column name from the CSV file.index: The index of the phrase in the CSV file.speakID: The ID of the speaker (0-9).stableVal: The stability value (10-50).
For example, part1_Base_0_0_style10.mp3 corresponds to the Base column in phrases_part1.csv, 0 index row, 0 speaker, and style10 (stability 0.1).