Spelling and NUmbers Voice database

70 Last view: 2023-11-10

Spelling and NUmbers Voice database

SNUV

http://pelcra.pl/res/spoken/snuv

ID:

513 SNUV (Spelling and NUmbers Voice database) is a spelling and number and recognition speech database containing over 220 hours of recordings of Polish speakers reading numbers and spelling words, recorded in 22050kHz, 16-bit *.wav files. 210 different participants were paid to produce a sample of their speech through an online spoken data collection platform. Written representation of the recordings is provided with the original sound files. The envisaged application of this resource is to enable the creation of automatic speech recognition (ASR) tools that allow users to spell out words and numbers to be recognized. SNUV has been released under a CC-BY license and cen be used for both academic and commercial purposes free of charge.

You don’t have the permission to edit this resource.

DistributionAvailability

Available - Restricted Use

Start date: 07/12/2012

Licence

CC - BY

Restrictions: Attribution

Download location: hidden

Distribution Access/Medium: Downloadable

Licensors:

Piotr Pęzik

Distribution rights holders:

IPR Holder

Contact Persons

text
audio

Monolingual text corpusLanguages

Polish

Language Script: Latn

Linguality

Linguality type: Monolingual

Size

220 Hours

704,625 Words

Character encoding

UTF - 8 (704,625 Words)

Domains

general (704,625 Words)

Modalities

Records of speakers reading out numbers and spelling words.

Spoken Language

AnnotationSpeech Annotation - Sound To Text Alignment

StandOff: False

Segmentation level: Utterance

Format: text/plain

Annotation Mode: Manual (All personal information has been anonymised.)

Start date: 02/01/2012

End date: 06/12/2012

Size: 704,625 Words

Geographic coverage

Poland (704,625 Words)

Creation

Creation mode details: Recordings of Polish speakers reading out numbers and spellings, time-aligned on the utterance level.

Creation mode: Manual

Creation Tools

SNUV Web application (http://snuv.pl)

Monolingual audio corpusLanguages

Polish (704,625 Files)

Linguality

Linguality type: Monolingual

Size

34 Gb

704,625 Words

Audio duration

220 Hours

Domains

general (704,625 Words)

Modalities

Recordings of Polish speakers reading out numbers and spellings, time-aligned on the utterance level. (704,625 Words)

Spoken Language (704,625 Words)

Classification

(704,625 Words)

Audio genre: Speech

Content

Speech items: Isolated Digits, Isolated Words, Natural Numbers

Noise Level: Low

Setting

Naturality: Prompted

Conversational type: Monologue

Interactivity: Other

Audio Formatsaudio/wav (704,625 Words)

Compression: False

Recording quality: Medium

Quantization: 16

Number of tracks: 1

Sampling rate: 22050

Signal encoding: LinearPCM

Geographic coverage

Poland (704,625 Words)

RecordingRecorders

VOICE LAB sp. z o.o.

Recording environment: Other

Recording device type: Hard Disk

Capture

Capturing device type details: The utterances were captured using an external microphone.

Capturing device type: Close Talk Microphone

Capturing environment: Complex

Person SourceSet

Age range end: 69

Age range start: 11

Geographic distribution of persons: Various regions across Poland.

Origin of persons: Native

Sex of persons: Mixed

Creation

Creation mode details: Recordings of Polish speakers reading out prompted numbers and spelling words, time-aligned on the utterance level.

Creation mode: Manual

Creation Tools

SNUV Web application (http://snuv.pl)

Resource Creation

Resource Creator

Piotr Pęzik

University of Łódź

Tomasz Szwelnik

Funding Project

Central and South-East European Resources (CESAR_PL - MNiSW 2139/CIP2007-2011/2)

URL: http://en.kpk.gov.pl...

Funding Type: National Funds

Funder: Polish Ministry of Science and Higher Education

Funding Country: Poland

Project duration: 01/02/2011 - 31/01/2013

Central and South-East European Resources (CESAR - 271022)

URL: http://www.meta-net....

Funding Type: Eu Funds

Funder: DG INFSO of the European Commission

Funding Country: European Union

Project duration: 01/02/2011 - 31/01/2013

Metadata

Created: 30/06/2012

Metadata Language: English (en)

Metadata Creator

Maciej Buczek

Piotr Pęzik

Version

Version: 1.0

Revision: compilation of the database

Last Updated: 07/12/2012

ValidationValidated (704,625 Words)

Type of Validation: Formal

Validation Mode: Automatic

Extent: Full

Validation Tools:

xmllint

Validator

Łukasz Dróżdż

Piotr Pęzik

Documentation

http://pelcra.pl/res...

People who looked at this resource also viewed the following:

Resources from the same project

Resources from the same creators