Hi,KIA: A Speech Emotion Recognition Dataset for Wake-Up Words

TL;DR

This paper introduces Hi, KIA, a Korean wake-up word dataset with emotional labels and provides baseline models for speech emotion recognition, addressing the lack of emotional data in wake-up word datasets.

Contribution

The paper presents a new emotional wake-up word dataset and baseline classification models, filling a gap in speech emotion recognition research for wake-up commands.

Findings

The dataset contains 488 emotional utterances from 8 speakers.

Baseline models achieve preliminary emotion recognition results.

The transfer-learning model outperforms traditional features.

Abstract

Wake-up words (WUW) is a short sentence used to activate a speech recognition system to receive the user's speech input. WUW utterances include not only the lexical information for waking up the system but also non-lexical information such as speaker identity or emotion. In particular, recognizing the user's emotional state may elaborate the voice communication. However, there is few dataset where the emotional state of the WUW utterances is labeled. In this paper, we introduce Hi, KIA, a new WUW dataset which consists of 488 Korean accent emotional utterances collected from four male and four female speakers and each of utterances is labeled with four emotional states including anger, happy, sad, or neutral. We present the step-by-step procedure to build the dataset, covering scenario selection, post-processing, and human validation for label agreement. Also, we provide two classification models for WUW speech emotion recognition using the dataset. One is based on traditional hand-craft features and the other is a transfer-learning approach using a pre-trained neural network. These classification models could be used as benchmarks in further research.