Results of tests on speech intelligibility in reverberant conditions

Description

The dataset contains the results of tests that aimed to provide a relationship between the rate of speech (RoS) and reverberation conditions characterized by the Speech Transmission Index (STI).

The subject’s task was to repeat the sentence and assess the rate of speech (appropriate or too fast). The supervisor verified the correctness of the response. If the answer was incorrect, the factor of slowing down was increased by 20%. A single test consisted of 13 trials; the first three trials were designed as a pre-test to familiarize the test subject with the sounds they needed to recognize.

The sentences had a fixed structure and were grammatically correct, although they have little contextual content. Each sentence consisted of five words: name, verb, numeral, adjective, and noun. An example sentence looked like this: “Tomasz nosi pięć dobrych piłek” (Thomas carries five good balls). The individual words forming the utterances were selected at random.

The tests used six impulse responses recorded at locations with different STI values. Response No. 3 had the lowest STI value (of 0.34), while response No. 6 had the highest value (of 0.76).

A total of 15 participants took part in the listening tests (3 females, 12 males). Their ages ranged between 24 and 42 years. All participants had previous hearing tests that revealed no impairments.

Flowchart of the sentence formation process

Dataset file

dataset.zip

154.4 kB, S3 ETag 2e57c3d12ba0f39914f84ef002a0e979-1, downloads: 20

The file hash is calculated from the formula
hexmd5(md5(part1)+md5(part2)+...)-{parts_count} where a single part of the file is 512 MB in size.

Example script for calculation:
https://github.com/antespi/s3md5

download

File details

License:: open in new tab

CC BY

Attribution
Raw data:: Data contained in dataset was not processed.

Details

Year of publication:

2021

Verification date:

2021-09-22

Dataset language:

English

Fields of science:

information and communication technology (Engineering and Technology)

DOI:

10.34808/b0yr-cm79

Verified by:

Gdańsk University of Technology

Keywords

References

publication Acoustic Sensing Analytics Applied to Speech in Reverberation Conditions

Cite as

Authors

seen 101 times

Search