Currently In Beta

Bengali Automatic Speech Recognition Dataset


Download Dataset    
About the dataset

Bangla Automatic Speech Recognition (ASR) dataset with 196k utterances.
The data set consists of wave files, and a TSV file. The file utt_spk_text.tsv contains a FileID, anonymized UserID and the transcription of audio in the file. The data set has been manually quality checked, but there might still be errors. See LICENSE file for license information.

Copyright 2016, 2017, 2018 Google, Inc.