English Dialects Combined is a dataset prepared for audio model training, merging 11 source datasets. The dataset includes added metadata for gender and accent for each audio record. It is based on the work 'Open-source Multi-speaker Corpora of the English Accents in the British Isles' presented at LREC 2020.
Use Cases
- Train accent classification models based on the described accent metadata.
- Develop multi-speaker speech synthesis systems based on the combined multi-speaker audio corpora.
- Benchmark speech recognition robustness across different English dialects based on the combined dataset goal.
- Analyze acoustic features correlated with speaker gender based on the provided gender metadata.
Strengths
- Combines 11 distinct source datasets, suggesting a broad base of audio samples.
- Includes added metadata fields for accent and gender for each record.
Limitations
- The specific number of rows, audio files, and total size are unknown.
- Column-level documentation is absent; field semantics must be inferred after download.
- Data may reflect geographic bias inherent to the source datasets focused on accents in the British Isles.
Provenance
- Source
- Combination of 11 datasets, originally from 🤗 ylacombe/english_dialects.
- Collection Method
- Merged from multiple source datasets with added metadata.
- Freshness
- Last updated 2026-04-30 14:54:13; freshness should be verified.
- Geography
- Likely focuses on English accents in the British Isles, based on the cited paper title.