British English Speech Recordings From 349 Native Speakers

Encompassing 9.5 hours of valid British English speech data from 349 native English speakers. Each speaker recorded about 50 sentences across multiple fields such as car, home, and voice assistant in a quiet environment.

Use Cases

• Train an automatic speech recognition model on 9.5 hours of British English audio from 349 speakers.
• Analyze speaker-specific phonetic variations using recordings from English locals across different content fields.
• Develop a voice assistant wake-word model using the 'voice assistant' field recordings from native speakers.

Strengths

• Data from 349 native English speakers provides a substantial sample of British accents.
• 9.5 hours of valid, transcribed speech data for model training.
• Recordings were made in a quiet environment, ensuring good audio quality.

Limitations

• The dataset is described as a sample of a larger paid dataset, indicating limited scope.
• No information is provided on the distribution of sentences across the mentioned fields (car, home, voice assistant).
• The sample size of 349 speakers may not capture the full diversity of British English dialects.

Provenance

Source

Nexdata on Hugging Face.

Collection Method

Recorded speech from native English speakers via mobile phone guiding.

Freshness

Last updated on 2025-04-23.

Geography

Speakers are English locals, indicating United Kingdom origin.