Hungarian Phone Speech Call Center Database

66 Last view: 2021-01-05

Hungarian Phone Speech Call Center Database

hu-MTUBA

http://alpha.tmit.bme.hu/speech/mtuba.php

ID:

209 The Hungarian Phone Speech Call Center Database is a telephone speech database containing discourses between the operators of a service provider company and its clients. Ortrhographc transcription is provided. Emotions are also labelled. A derivative of this database called Hungarian Speech Emotion Database is also available from META-SHARE, with free academic use.

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

DistributionAvailability

Available - Restricted Use

Start date: 02/07/2012

Licence

MS - NC - No ReD - ND - FF

Restrictions: No Derivatives, No Redistribution

Fee: 8,000 EUR

Distribution Access/Medium: CD - ROM

Attribution Details: A derivative of this database called Hungarian Speech Emotion Database is also available from META-SHARE, with free academic use, however, without ortographic transcription.

MS - C - No ReD - ND - FF

Restrictions: No Derivatives, No Redistribution

Fee: 12,000 EUR

Distribution Access/Medium: CD - ROM

Attribution Details: A derivative of this database called Hungarian Speech Emotion Database is also available from META-SHARE, with free academic use, however, without ortographic transcription.

IPR Holder

Klára Vicsi

Contact Person

Klára Vicsi

text
audio

Monolingual text corpusLanguages

Hungarian

Linguality

Linguality type: Monolingual

Size

1,038 Utterances

Creation

Creation mode: Manual

Original Sources

research

Monolingual audio corpusLanguages

Hungarian (1,038 Utterances)

Linguality

Linguality type: Monolingual

Size

1,038 Utterances

Size

1.8 Gb

Size

3.5 Hours

AnnotationSemantic Annotation - Emotions

Annotated elements: Other

Segmentation level: Phrase

Annotation Mode: Manual (annotation based on listening)

Annotation Tools:

Praat

Start date: 01/01/2009

End date: 29/02/2012

Content

Noise Level: Medium

Audio Formatswave/audio (3.5 Hours)

Compression: False

Recording quality: Medium

Quantization: 8

Number of tracks: 1

Sampling rate: 8000

Signal encoding: A Law

CapturePerson SourceSet

Origin of persons: Native

Age of persons: Adult

Sex of persons: Mixed

Number of persons: 1038

Dialect accent of persons: varied

Age range end: 99

Hearing impairment of persons: No

Number of trained speakers: 0

Age range start: 18

Speaking impairment of persons: No

Resource Creation

Resource Creator

Klára Vicsi

Creation lasted: 01/01/2009 - 29/02/2012

Metadata

Created: 02/07/2012

Last Updated: 02/07/2012

Source: CESAR

Metadata Creator

György Szaszák

Version

Version: 2.0

Last Updated: 29/02/2012

Usage

Foreseen UseNlp Applications

Use NLP Specific: Emotion Recognition, Person Recognition, Speech Analysis, Speech Recognition, Spoken Dialogue Systems

Human UseActual Use - Human Use

Use NLP Specific: Emotion Recognition, Speech Recognition

People who looked at this resource also viewed the following:

Resources from the same creators