Emotion Markup Language 1 EmotionML: Overview on IRs: Content Status of EmotionML Spec Overview potentially non-implemented features Overview Implementers ALMA C# library EMLPy EMO20Q Gtrace Mary TTS Nviso Speechalyzer Wasabi
Jan 24, 2016
Emotion Markup Language 1
EmotionML: Overview on IRs: Content
Status of EmotionML SpecOverview potentially non-implemented featuresOverview Implementers
ALMA
C# library
EMLPy
EMO20Q
Gtrace
Mary TTS
Nviso
Speechalyzer
Wasabi
Emotion Markup Language 2
EmotionML: Overview on Implementation Reports
Status: Last call working draftCollecting Implementation Reports and answering public feedbackBy now 9 IRs were submitted, one is announced (TU-Munich) but will come too late (end of year)The core of the spec was implemented, but some elements are in danger (following slide)We’re in the process of waiting for extensions to implement missing featuresThere’s also a common journal publication planned under the lead of Edmon Begoli
Emotion Markup Language 3
EmotionML: Features in danger
There’s two implementations, that support action-tendency and appraisal-set, but don’t check the names against the vocab, which is required.
Implementers promised to deliver till end of month
There’s only one implementation so far for the time-stuff taken from EMMA, i.e. start, end, duration, time-ref-uri, offset-to-start.
Because we declared these “at risk”, it should be ok to keep them in the spec.
There are two implementations that use “media-type” (416), but both don’t check if it’s a valid mime-type, which is an required assertion (417).
The implementers are notified and try t implement this, although it’s unclear which mime-type vocab to check
Emotion Markup Language 4
EmotionML: IR submitting institutions
Telekom Innovation Laboratories
Emotion Markup Language 5
EmotionML IR: Alma
DFKI
http://www.dfki.de/~gebhard/alma/
ALMA: A Layered Model of Affect.
is a computational model for the real-time simulation of three basic affect types
The ALMA system is an affect producing system.
It outputs EmotionML
Emotion Markup Language 6
EmotionML IR: C# Library
Univ. of Chemnitz
https://github.com/gfobe/EmotionML-Lib-CSharp
Master’s thesis at the Univ. of Chemnitz
Open source project at Github
Used in “Smiley ontology”
EmotionML C# library used to describe Emoticons in IRC chat protocol services
Emotion Markup Language 7
EmotionML IR: EMLPy
Oak Ridge National Laboratory/University of Tennessee
https://github.com/ebegoli/EMLPy
a Python based library for generation of EmotionML compliant documents.
Open source with Github
ultimately (and distantly) using EmotionML to help autistic children with alternative representations of emotional content in the material.
Emotion Markup Language 8
EmotionML IR: EMO20Q
USC-SAIL
http://sail.usc.edu/emo20q/
a experimental framework for studying how people describe emotions in language
and how computers can simulate this type of verbal behavior.
Open source
Emotion Markup Language 9
EmotionML IR: Gtrace
Queen’s Univ. Belfast
http://mary.dfki.de/
Software to trace emotional expression in videos.
The system currently implements tracing for category and dimensional descriptors.
Emotion Markup Language 10
EmotionML IR: Mary TTS
DFKI
http://mary.dfki.de/
MARY is an open-source, multilingual Text-to-Speech Synthesis platform that includes modules for expressive speech synthesis.
Particularly the support for both categorical and dimensional representations of emotions is important to its expressive speech synthesis system MARY TTS.
Emotion Markup Language 11
EmotionML IR: NViso
nViso
http://nviso.ch
nViso 3D Facial Imaging API is an online service for recognition emotions depicted through facial expressions in still images and videos.
The focus of this implementation of EmotionML is on using the media type and URI time for video.
Emotion Markup Language 12
EmotionML IR: Speechalyzer
Deutsche Telekom Laboratories
https://github.com/dtag-dbu/speechalyzer
Analysis / annotation / transcription tool “Speechalyzer”
Open Source project in Github
Can be used to rapidly judge large number of audio files emotionally.
Automatic classification integrated.
Uses EmotionML as exchange format.
Emotion Markup Language 13
EmotionML IR: Wasabi
Univ. of Freiburg
https://github.com/CBA2011
WASABI architecture for affect simulation
Applied to the articulated communicator Max of the Univ. of Bielefeld
To let the WASABI architecture seamlessly interface with other software modules, a standard interface such as EmotionML is indispensable.