Infant and Toddler Crying Audio Database

1. Product Overview

The infant and toddler crying audio database comprises 68 hours of audio, totaling 2,549 recordings. The recordings were made in China, involving 544 children. All recordings were conducted in quiet indoor environments using high-fidelity equipment, Android smartphones, and iOS smartphones, with an audio sampling rate of 16 kHz and a bit depth of 16 bits. This database is suitable for research applications such as voiceprint/voice recognition and signal detection.

2. Speaker Information

Speaker information includes task ID, gender, and age range. Each participant receives one task ID and may only take part in a single recording session. The age range covers infants and toddlers aged 2 to 6 years. The database contains a total of 544 speakers, including 281 males and 263 females (51.7% male, 48.3% female).

3. Recording Environment

The recordings were captured in quiet indoor settings with minimal background noise. Recording equipment included microphones, Android smartphones, and iOS smartphones.

4. Data File Directory and Naming Conventions

The data directory structure is as follows: the DOC directory (containing text documentation) includes the audio list file wav_list.txt and the speaker information file spkrinfo.xlsx; the SPEECHDATA directory (data folder) organizes audio files into subdirectories corresponding to individual speakers.

Directory naming convention: //, e.g., SPEECHDATA/0364/P0364W0001.wav. File naming convention: , e.g., P0364W0001.wav.

5. Sample Data Download

Click to download the sample dataset (Infant and Toddler Crying Audio Database.zip, approximately 1.9 MB, including DOC documentation and example audio files from the SPEECHDATA folder)