SUBESCO is an audio-only emotional speech corpus of 7000 sentence-level utterances of the Bangla language. 20 professional actors (10 males and 10 females) participated in the recordings of 10 sentences for 7 target emotions. The emotions are Anger, Disgust, Fear, Happiness, Neutral, Sadness and Surprise. Total duration of the corpus is 7 hours 40 min 40 sec. Total size of the dataset is 2.03 GB. The dataset was evaluated by 50 raters (25 males, 25 females). Human perception test achieved a raw accuracy of 71%. All the details relating to creation, evaluation and analysis of SUBESCO have been described in the corresponding journal paper which has been published in Plos One. https://doi.org/10.1371/journal.pone.0250173 File naming convention: Each audio file in the dataset has a unique name. There are eight parts in the file name where all the parts are connected by underscores. The order of all the parts is organized as: Gender-Speaker's serial number-Speaker's name-Unit of recording-Unit number- Emotion name- Repeating number and the File format. For example, the filename F_02_MONIKA_S_1_NEUTRAL_5.wav refers to: • Female speaker (F) • Speaker number(02) • Speaker's name (MONIKA) • Sentence number (S_1) • Emotional state (NEUTRAL) • Take number (5) Contact Information If you would like further information about the SUBESCO, or if you experience any issues downloading files, please contact at sadia-cse@sust.edu.

This database was created as a part of PhD thesis project of the author Sadia Sultana. It was designed and developed by the author in the Department of Computer Science and Engineering of Shahjalal University of Science and Technology. Financial grant was supported by the university. If you use the dataset please cite SUBESCO and the corresponding academic journal publication in Plos One.