ATC Voice Data Annotators Needed

We are looking for volunteers willing to annotate ATC voice data.
If interested, please drop e-mail to This email address is being protected from spambots. You need JavaScript enabled to view it.

I’m really excited to say that OpenSky Network is part of a project called “AuTomatic COllection and processing of voice data from Air-Traffic Communications” (ATCO2). The project aims at developing a unique platform allowing to collect, organize and pre-process ATC voice data and to make it available to public use.
The project is financed by Clean Sky, the largest European research programme, funded by the EU’s Horizon 2020 programme.

Why it matters?
One possible way: Airbus, one of the project members, is expecting that automatic transcription algorithms developed by the project, can be used to display ATC information as text in cockpit. That will allow pilots to return to it later on without the need of repeating the clearances by ATC. Obvious outcome of it is reduced workload for all related parties.

One can think many more examples how the availability of transcribed ATC voice data could be used to improve safety and efficiency of ATM.

It's also will provide invaluable source of information to ATM related research. Having transcribed ATC voice data together with crowed sourced surveillance data will give researchers unique insight into everyday operations like never before. This, in turn, will open up new possibilities for ATM efficiency, safety and security related research.

What's the approach?
As a first step, we aim to deploy community-based VHF receiver network, that continuously collects voice data and sends it to OpenSky, where it will be stored for further processing.
Sounds familiar? Well, it should! Basically, we will expand OpenSky Network dataset to include voice data.

As a second step, ATCO2 will implement automatic speech recognition and transcription algorithms. As with any other machine learning algorithm, people are needed to make sure that the machine gets it right.
And that is the place, where we need You. There are two main tasks to be carried out:
• annotate voice data to text so the ML algorithm has data to build its knowledge base and
• check if the automatic transcription algorithm got it right.

Are you up for the task?
If yes, make yourself seen by sending an e-mail to me: This email address is being protected from spambots. You need JavaScript enabled to view it.

What’s in it for you?
Well, the incentive package will be developed during the project, but probably it will include access to the data (there will be some limitations/prerequisites to other) and why not, maybe even free VHF receivers to become a feeder. And off course, You will get the joy of improving aviation safety, efficiency and security.

List of project partners can be found here: www.atco2.org/partners .
In addition to research partners, Honyewell and Airbus are included from the industry side.