Please register below and you can start to help us validate the correctness of automatic transcripts of ATC communication. Your help free of charge is very appreciated. In the following chart you can see the number of recordings that were processed in the last couple days. Click the button below to see the long story.
The ATCO2 project is aiming at collection of open data in the ATC domain. The goal of the project is to support a large community of people interested in anything related to air (-planes, -traffic, -communication, ...).
The air traffic communication speech is collected by a group of enthusiasts via VHF channel. Another group is collecting the ADS-B data. Both data streams are sent to the Opensky Network servers. Well known hubs of the VHF data is LiveATC, and Opensky Network or FlightRadar24 for the ADS-B data. We decided to push the Opensky Network on another level by merging the VHF and ADS-B data streams together. So you will not only be able to track the airplane but also to listen to the voice communication and read the transcript.
To do so, we need the help of the community. Automatic transcription of ATC data is tough. We know how to do it thanks to our university partners in ATCO2. The missing puzzle is the data. We need tens of hours of human transcribed data. And here is where you enter the game. Please help us and the whole community of fans and enthusiasts to let it happen.
As you do it voluntarily and free of charge you may ask, what will you or the community get out of it? Three things:
You should follow the annotation manual during annotations.
So far we are interested in English language. But we are open to support other languages. Feel free to help us by being a VHF data feeder or the transcriber. If you are interested in a particular airport or English accent (if you are French / Czech you can cope with Frenglish / Czenglish much easier than others for example) please select your preferences below.
This service is being developed under the ATCO2 project. We are behind this online service for audio data annotations. The audio data comes from feeders through the OpenSky Network while our research partners (BUT, IDIAP, SAARLAND) are providing automatics speech processing technologies.
Please remember that there may be limitations set by your own national legislation concerning the recording of air traffic speech data. By joining our community of annotators you engage in making the opportune verifications and engage not to infringe any applicable rule.
In case of doubt, you can contact us and we will give you support with the help of ATCO2 partners.
