Dataset · ylacombe
Speech recordings with various British Isles accents
ylacombe/english_dialects
31 hours of transcribed studio audio from 120 speakers of British and Irish dialects for TTS.
What's inside
It contains 31 hours of high-quality audio recordings of English sentences (from 10K to 100K samples) with text transcripts. The recordings were made with 120 volunteers who are native speakers of accents from Southern, Central, and Northern England, as well as Wales, Scotland, and Ireland.
What you can do with it
- Training TTS models and voice assistants with support for regional British accents
- Linguistic analysis of phonetic features of British Isles dialects
- Tasks
- Text → speech, Text → audio
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 1.2K
- Likes
- 40
Models trained on it
- futo-org/asr4all-lAutomatic speech recognition 794
- futo-org/asr4all-sAutomatic speech recognition 592
- futo-org/asr4all-mAutomatic speech recognition 458
- futo-org/asr4all-l-plusAutomatic speech recognition 208
- futo-org/asr4all-m-plusAutomatic speech recognition 96
- futo-org/asr4all-s-plusAutomatic speech recognition 84
- Vira21/Whisper-Base-KhmerV2Automatic speech recognition 56
- MohammadDaleen/speecht5_finetuned_english_dialectsText → speech 16
How to load it
from datasets import load_dataset
ds = load_dataset("ylacombe/english_dialects")