Malay Speech Audio Samples From Mobile Phone Recordings

Aggregating a sample of 134 hours of Malay speech audio recorded by 156 native speakers in a quiet environment. It includes approximately 450 sentences per speaker, covering categories like economy, entertainment, news, and numbers.

Use Cases

• Train an automatic speech recognition (ASR) model on 134 hours of native Malay speaker audio recordings.
• Analyze speech patterns and acoustic features across different content categories such as economy, entertainment, and news.
• Develop speaker identification or verification models using audio from 156 distinct speakers.
• Build a text-to-speech (TTS) frontend by studying the phonetic and prosodic characteristics of the recorded oral language and numbers.

Strengths

• Contains 134 hours of effective audio data.
• Recordings were made by 156 native Malay speakers.
• Content covers multiple defined categories including economy, entertainment, and news.
• Recordings were conducted in a quiet environment to ensure audio clarity.

Limitations

• The dataset is described as a sample of a larger paid dataset, so it may not be the complete collection.
• The exact number of audio files or rows is not specified.
• Specific audio quality metrics, file formats, and transcription details are not provided.

Provenance

Source

Nexdata

Collection Method

Recorded by native Malay speakers using mobile phones in a quiet environment.

Freshness

Last updated on 2025-04-23.