Detection of Baby Voice and its Application Using Speech Recognition System and Fundamental Frequency Analysis

摘要

We propose a method for detecting a baby voice using a speech recognition system and fundamental frequency analysis. We propose the following two conditions for recognizing a sound form segment of a baby voice. Condition 1: The word reliability for a sound form segment obtained by using Julius is under a threshold, Condition 2: For a certain time period, the fundamental frequency of the sound form segment changes by another threshold or over. When at least one of the above two conditions is met, the sound form segment is judged as coming from a baby voice. We successfully applied the proposed method to pattern recognition of a baby's emotion.