Spanish Speech Recordings From Mobile Phones

Aggregating 9.9 hours of Spanish speech recordings from 343 native speakers in Spain, Mexico, and Argentina. Each speaker contributed 50 sentences, recorded in a quiet environment using mainstream Android phones and iPhones. All audio has been manually transcribed with high accuracy.

Use Cases

• Train an automatic speech recognition (ASR) model on manually transcribed Spanish audio from 343 speakers.
• Analyze phonetic or accent variations using recordings from speakers in Spain, Mexico, and Argentina.
• Fine-tune a voice activity detection model on 9.9 hours of quiet-environment recordings from mobile devices.
• Develop a speaker diarization system using the 50-sentence-per-speaker structure of the recordings.

Strengths

• Data from 343 native Spanish speakers provides a multi-speaker foundation.
• 9.9 hours of total audio offers a substantial volume for speech tasks.
• Manual transcription ensures high-accuracy text labels for the audio.
• Recordings originate from three distinct Spanish-speaking regions: Spain, Mexico, and Argentina.

Limitations

• The dataset is described as a sample of a larger paid dataset, implying limited scope.
• Specific audio file formats, sampling rates, and metadata columns are not provided.
• The quiet recording environment may not represent real-world noisy conditions.

Provenance

Source

Nexdata

Collection Method

Recordings collected from 343 native speakers using mainstream Android phones and iPhones in a quiet environment.

Time Range

null

Freshness

null

Geography

Spain, Mexico, Argentina