Open Speech and Language Resources



Large Bengali ASR training data set

Identifier: SLR53

Summary: Bengali ASR training data set containing ~196K utterances.

Category: Speech

License: Attribution-ShareAlike 4.0 International

Downloads (use a mirror closer to you):
about.html [1.0K]   (Information about the data set )   Mirrors: [US]   [EU]   [CN]  
LICENSE [20K]   (License information for the data set )   Mirrors: [US]   [EU]   [CN]  
utt_spk_text.tsv [15M]   (All utterances in the data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_0.zip [919M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_1.zip [906M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_2.zip [921M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_3.zip [918M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_4.zip [908M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_5.zip [932M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_6.zip [900M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_7.zip [927M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_8.zip [927M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_9.zip [906M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_a.zip [900M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_b.zip [910M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_c.zip [897M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_d.zip [914M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_e.zip [922M]   (Data set )   Mirrors: [US]   [EU]   [CN]  
asr_bengali_f.zip [917M]   (Data set )   Mirrors: [US]   [EU]   [CN]  

About this resource:

This data set contains transcribed audio data for Bengali. 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.txt file for license information.

Copyright 2016, 2017, 2018 Google, Inc.

If you use this data in publications, please cite it as follows:

  @inproceedings{kjartansson-etal-sltu2018,
    title = {{Crowd-Sourced Speech Corpora for Javanese, Sundanese,  Sinhala, Nepali, and Bangladeshi Bengali}},
    author = {Oddur Kjartansson and Supheakmungkol Sarin and Knot Pipatsrisawat and Martin Jansche and Linne Ha},
    booktitle = {Proc. The 6th Intl. Workshop on Spoken Language Technologies for Under-Resourced Languages (SLTU)},
    year  = {2018},
    address = {Gurugram, India},
    month = aug,
    pages = {52--55},
    URL   = {http://dx.doi.org/10.21437/SLTU.2018-11}
  }