Machine Learning of Speech Recognition Models for Controller Assistance

Project Information

MALORCA

Grant agreement ID: 698824

DOI

10.3030/698824

Project closed

EC signature date 30 March 2016

Start date 1 April 2016

End date 31 March 2018

Funded under

• SOCIETAL CHALLENGES - Smart, Green And Integrated Transport

Total cost

€ 805 587,50

EU contribution

€ 538 103,75

Coordinated by

DEUTSCHES ZENTRUM FUR LUFT - UND RAUMFAHRT EV
Germany

CORDIS provides links to public deliverables and publications of HORIZON projects.

Links to deliverables and publications from FP7 projects, as well as links to some specific result types such as dataset and software, are dynamically retrieved from OpenAIRE .

Deliverables

Documents, reports (4)

Gap Analysis Report (T0+24)

The last step of WP5 and also of the project will be the gap analysis with respect to achieved results.

System Requirement Document

The document is the output of task 1.2 (Consolidation of system requirements). It will define the system requirements with special consideration of interfaces, data content and formats of the data preparation, and data privacy issues in connection to WP2. Functional requirements will be defined to specify clearly what a system is su-posed to accomplish. The performance requirements will be elaborated here.

Publishable Project Results final report

The Final Project Report includes a publishable summary of the entire project. The report is the input for the Project Closeout Meeting. This deliverable will summarize the project scope and objectives, the achieved results and main conclusions, the performed communication and dissemination actions, and the exploitation and follow-up activities proposed for the next stage of the R&I lifecycle. It will include a self-assessment of the achieved TRL of the MALORCA unsupervised learning component. Furthermore it will include a short summary of the gap analysis and the stakeholder feedback from the workshops and the proof of concept experiments.

Operational Concept Description

The Operation Concept Description will define the user requirements. Based on the operational experience of ANSPs, they will provide the set of use cases in connection to voice recognition application in ATM environment with the main focus on automatic learning of speech recognition parameters. Those uses cases will be evolved from the current operational practise and their identified limitations. This task also identifies the current operating environment from the view of the end users (controllers). Identified use cases will serve as a basis for the user requirements which will be imbedded in the Operational Concept Document

Websites, patent fillings, videos etc. (1)

Report on Stakeholder Workshop Results

Austro Control and ANS CR, are the first test sites who will explore proposed solutions. Both partners plan to organize at least two workshops (within the project duration) to demonstrate the developed tools for all interested stakeholders from other ANSPs and also for interested suppliers. A first stakeholder workshop will be conducted after 15 months resulting in a draft of this deliverable. After a 23 months a second workshop is planned which will be the input into this deliverable.

Publications

Conference proceedings (3)

A CONTEXT-AWARE SPEECH RECOGNITION AND UNDERSTANDING SYSTEM FOR AIR TRAFFIC CONTROL DOMAIN

Author(s): Youssef Oualil, Dietrich Klakow, Gyorgy Szaszak, Ajay Srinivasamurthy, Hartmut Helmke, Petr Motlicek
Published in: Automatic Speech Recognition and Understanding Workshop, 2017
Publisher: Automatic Speech Recognition and Understanding Workshop

Machine Learning of Controller Command Prediction Models from Recorded Radar Data and Controller Speech Utterances

Author(s): M. Kleinert, H. Helmke, G. Siol, H. Ehr, M. Finke, Y. Oualil, A. Srinivasamurthy
Published in: SESAR Innovation Days in Belgrade, 2017, Page(s) 8
Publisher: SESAR Innovation Days

Semi-supervised Learning with Semantic Knowledge Extraction for Improved Speech Recognition in Air Traffic Control

Author(s): Ajay Srinivasamurthy, Petr Motlicek, Ivan Himawan, Gyorgy Szaszak, Youssef Oualil, Hartmut Helmke
Published in: 2017
Publisher: Interspeech 2017

Other (1)

MALORCA: Learning of Controllers' Behaviour from Recorded Radar Data and Speech Utterances

Author(s): Hartmut Helmke, Petr Motlicek, Ivan Himawan, Oliver Ohneiser, Christian Windisch, Christian Kern, Aneta Cerna, Matej Nesvadba, Youssef Oualil, Dietrich Klakow, György Szaszak
Published in: 6th SESAR INNOVATION DAYS, 2016
Publisher: SJU