File size: 2,624 Bytes
c1a5f71 175d948 4be2171 175d948 916ad76 ccf12f1 ce6a0c4 672a5ee 175d948 f71436c ccf12f1 c1a5f71 4473e2b b86628c 4473e2b 115c2d5 72f783a d396691 115c2d5 3f2d913 115c2d5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 |
---
language:
- en
license: cc0-1.0
source_datasets:
- mozilla-foundation/common_voice_14_0
task_categories:
- text-to-audio
- automatic-speech-recognition
- audio-to-audio
- audio-classification
dataset_info:
features:
- name: audio
dtype: audio
- name: speaker_id
dtype: string
- name: transcript
dtype: string
- name: accent
dtype: string
- name: duration
dtype: float64
- name: age
dtype: string
- name: gender
dtype: string
splits:
- name: test
num_bytes: 496943021.995
num_examples: 5455
- name: train
num_bytes: 53758082721.361
num_examples: 572159
- name: val
num_bytes: 373541300.088
num_examples: 4111
download_size: 47602304610
dataset_size: 54628567043.444
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: val
path: data/val-*
- split: train
path: data/train-*
---
# Important notice
### !!! Please use [V2 version](https://huggingface.co/datasets/MushanW/GLOBE_V2) version as this version has abnormal voice volume issue.
# Globe
The full paper can be accessed here: [arXiv](https://arxiv.org/abs/2406.14875)
An online demo can be accessed here: [Github](https://globecorpus.github.io/)
## Abstract
This paper introduces GLOBE, a high-quality English corpus with worldwide accents, specifically designed to address the limitations of current zero-shot speaker adaptive Text-to-Speech (TTS) systems that exhibit poor generalizability in adapting to speakers with accents. Compared to commonly used English corpora, such as LibriTTS and VCTK, GLOBE is unique in its inclusion of utterances from 23,519 speakers and covers 164 accents worldwide, along with detailed metadata for these speakers. Compared to its original corpus, i.e., Common Voice, GLOBE significantly improves the quality of the speech data through rigorous filtering and enhancement processes, while also populating all missing speaker metadata. The final curated GLOBE corpus includes 535 hours of speech data at a 24 kHz sampling rate. Our benchmark results indicate that the speaker adaptive TTS model trained on the GLOBE corpus can synthesize speech with better speaker similarity and comparable naturalness than that trained on other popular corpora. We will release GLOBE publicly after acceptance.
## Citation
```
@misc{wang2024globe,
title={GLOBE: A High-quality English Corpus with Global Accents for Zero-shot Speaker Adaptive Text-to-Speech},
author={Wenbin Wang and Yang Song and Sanjay Jha},
year={2024},
eprint={2406.14875},
archivePrefix={arXiv},
}
``` |