BIIC-Podcast Overview

The BIIC-Podcast database is the collection of emotional speech recordings in Taiwanese Mandarin language. The audio was gathered from various audio-sharing platforms that provide creative commons (CC) licenses under CC-BY or CC-BYSA. To ensure diversity in the collection, we carefully selected topics (sports, lifestyle, business, music, and more), including monologues and conversations, and in various categories such as drama, interviews, casual conversations etc. We also maintained a balance between male and female speakers and limited the duration of one speaker to prevent speaker bias.

The raw audio recordings included in the BIIC-Podcast database have undergone all the stages of the intelligently controlled pipeline. This corpus is continuously collected in an ongoing process through the mentioned pipeline, which involves switching or tweaking some of the components to include language-specific emotional information.

The data set is publicly available and we encourage researchers to use it for data mining and testing their own SER models. For a full description of the data set, please refer to the following paper:

Shreya G. Upadhyay*, Woan-Shiuan Chien*, Bo-Hao Su, Lucas Goncalves, Ya-Tse Wu, Ali N. Salman, Carlos Busso and Chi-Chun Lee, “An Intelligent Infrastructure Toward Large Scale Naturalistic Affective Speech Corpora Collection,” in 2023 11th International Conference on Affective Computing and Intelligent Interaction (ACII). IEEE, 2023, pp. https://ieeexplore.ieee.org/#

If you use this data set, we request that you cite this paper in your work.

Release of the Corpus: Academic License

The corpus is now available under an Academic License (free of cost). Please download this pdf. The process requires your institution to sign the agreement. A couple of notes about this form:

Release of the Corpus: Commercial License

Companies interested in this corpus can obtain a commercial license from National Tsing Hua University.