1977 results found.
Mask Augsburg Speech Corpus (MASC)
Speech
Corpus,
Language Type:
Monolingual
Languages:
German
Availability:
From Owner
License:
Free within participation in Interspeech 2020 ComParE
Size:
10 hours Production Status:
Newly created-finished
Use:
Evaluation/Validation
-
Paper title:The INTERSPEECH 2020 Computational Paralinguistics Challenge: Elderly Emotion, Breathing & Masks
-
Paper track:13.15 The INTERSPEECH 2020 Computational Paralingu/Oral Presentation
-
Paper status:Accept Special Session
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Björn Schuller | Ulm State of Mind in Speech-elderly (USOMS-e) | /N |
| Main Contact | Björn Schuller | UCL Speech Breath Monitoring (UCL-SBM) | /N |
| Main Contact | Björn Schuller | Mask Augsburg Speech Corpus (MASC) | /N |
Documentation:
None
Action Game Speech Communication Corpus
Multimodal/Multimedia
Corpus,
Language Type:
Monolingual
Languages:
Japanese
Availability:
Public release in preparation
License:
Size:
3.0 GByte Production Status:
Newly created-finished
Use:
Dialogue
-
Paper title:Gaming Corpus for Studying Social Screams
-
Paper track:3.6 Social signal processing/Poster Presentation
-
Paper status:Accept - Poster
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Hiroki Mori | Action Game Speech Communication Corpus | /N |
Documentation:
None
Wall Street Journal (WSJ) Corpus
Speech
Corpus,
Language Type:
Monolingual
Languages:
English
Availability:
License:
LDC
Size:
None Production Status:
Existing-used
Use:
Machine Learning
-
Paper title:Multi-talker ASR for an unknown number of sources: Joint training of source counting, separation and ASR
-
Paper track:5.8 Source separation and computational auditory s/Oral Presentation
-
Paper status:Accept - Oral
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Thilo von Neumann | Wall Street Journal (WSJ) Corpus | /N |
Documentation:
None
The Suomi 24 Sentences Corpus
Written
Corpus,
Language Type:
Monolingual
Languages:
Finnish
Availability:
From Owner
License:
CC - BY - NC
Size:
4132665850 tokens Production Status:
Existing-used
Use:
Language Modelling
-
Paper title:FinChat: Corpus and evaluation setup for Finnish chat conversations on everyday topics
-
Paper track:12.19 Other topics in Spoken Language Processing: /Oral Presentation
-
Paper status:Accept - Poster
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Katri Leino | OpenSubtitles | /N |
| Main Contact | Katri Leino | The Suomi 24 Sentences Corpus | /N |
| Main Contact | Katri Leino | FinChat | /N |
Documentation:
None
ERATO Human-Robot Interaction Corpus
Multimodal/Multimedia
Corpus,
Language Type:
Monolingual
Languages:
Japanese
Availability:
Not Available
License:
Size:
130 sessions of interactions OtherProduction Status:
Newly created-in progress
Use:
Corpus Creation/Annotation
-
Paper title:Semi-supervised learning for character expression of spoken dialogue systems
-
Paper track:11.1 Spoken dialog systems/Oral Presentation
-
Paper status:Accept - Poster
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Kenta Yamamoto | ERATO Human-Robot Interaction Corpus | /N |
Documentation:
None
JVS (Japanese versatile speech) corpus
Speech
Corpus,
Language Type:
Monolingual
Languages:
Japanese
Availability:
Freely Available
License:
Size:
3.5 GByte Production Status:
Existing-used
Use:
Speech Synthesis
-
Paper title:Cross-lingual Text-To-Speech Synthesis via Domain Adaptation and Perceptual Similarity Regression in Speaker Space
-
Paper track:7.14 Cross-lingual and multilingual aspects in spe/Poster Presentation
-
Paper status:Accept - Poster
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Detai Xin | The LJ Speech Dataset | /N |
| Main Contact | Detai Xin | JSUT (Japanese speech corpus of Saruwatari-lab., University of Tokyo) | /N |
| Main Contact | Detai Xin | CSTR VCTK Corpus | /N |
| Main Contact | Detai Xin | JVS (Japanese versatile speech) corpus | /N |
Documentation:
English and Japanese documentation available
Wall Street Journal (WSJ) Corpus
,
Language Type:
Language Independent
Languages:
<Not Specified>
Availability:
License:
Size:
None Production Status:
Use:
-
Paper title:Raw Sign and Magnitude Spectra for Multi-head Acoustic Modelling
-
Paper track:8.1 Feature extraction and low-level feature model/Oral Presentation
-
Paper status:Accept - Oral
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Erfan Loweimi | Wall Street Journal (WSJ) Corpus | /N |
Documentation:
None
Düsseldorf Anxiety Corpus - Sustained Vowels
Speech
Corpus,
Language Type:
Language Independent
Languages:
<Not Specified>
Availability:
From Owner
License:
Unknown
Size:
3 hours Production Status:
Existing-used
Use:
Speech Recognition/Understanding
-
Paper title:An Evaluation of the Effect of Anxiety on Speech - Computational Prediction of Anxiety from Sustained Vowels
-
Paper track:3.3 Automatic analysis of speaker states/Oral Presentation
-
Paper status:Accept - Oral
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Alice Baird | Düsseldorf Anxiety Corpus - Sustained Vowels | /N |
Documentation:
None publicly available.
Wall Street Journal (WSJ) Corpus
Speech/Written
Corpus,
Language Type:
Monolingual
Languages:
English
Availability:
From Data Center(s)
License:
LDC
Size:
83.28 hours Production Status:
Existing-used
Use:
Speech Recognition/Understanding
-
Paper title:Do End-to-End Speech Recognition Models Care About Context?
-
Paper track:8.14 Other topics in Speech Recognition: Signal Pr/Poster Presentation
-
Paper status:Accept - Poster
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Lasse Borgholt | LibriSpeech ASR corpus | /N |
| Main Contact | Lasse Borgholt | Wall Street Journal (WSJ) Corpus | /N |
| Main Contact | Lasse Borgholt | TIMIT Acoustic-Phonetic Continuous Speech Corpus | /N |
Documentation:
None
Mandarin annotated spoken (MAS) corpus
Speech
Corpus,
Language Type:
Monolingual
Languages:
Mandarin Chinese
Availability:
From Data Center(s)
License:
Size:
25 GByte Production Status:
Existing-updated
Use:
Evaluation/Validation
-
Paper title:An Effective End-to-End Modeling Approach for Mispronunciation Detection
-
Paper track:10.2 Applications in education and learning (incl./Oral Presentation
-
Paper status:Accept - Oral
| Author Number | Name | Affiliation | Country |
|---|---|---|---|
| Main Contact | Tien-Hong Lo | Mandarin annotated spoken (MAS) corpus | /N |
Documentation:
1). Yes. 2). Mandarin and English. 3). Please contact us.