International Conference on Computer and Knowledge Engineering

Home / 12th International Conference on Computer and Knowledge Engineering

Robat-e-Beheshti: A Persian Wake Word Detection Dataset for Robotic Purposes

Authors :

Parisa Ahmadzadeh Raji¹ Yasser Shekofteh²

1- Faculty of Computer Science and Engineering, Shahid Beheshti University 2- Faculty of Computer Science and Engineering, Shahid Beheshti University

Keywords :

Speech Recognition،Wake Word،Convolutional Neural Network،Dataset،Transfer Learning

Abstract :

In this paper we explain a dataset that we collected for a wake word detection project which works on classifying audio data. The data is audio files in persian language. We collected 5738 audio files with different formats and a maximum length of 3 seconds. Then we changed the formats and their sample rates to the specific one. We have positive and negative examples for our dataset. Different audio recorders were used for recording the audios and most of the data gathered from 187 individuals and the other files collected from open source ShEMO dataset: a large-scale validated database for Persian speech emotion detection. In this paper we explain the process of collecting data, how they are organized in different files and also we explain the data analysis. This dataset will be free for academic usage with official request to us.