AI-based voice recognition has great potential to make technology more inclusive and enable millions of people access to services they are not able to use yet – be it in health, education or others. What is needed? Open voice data!

People and startups relying on voice-recognition technology face one problem alike: A lack of freely available voice data in their respective language to train artificial intelligence. Although many machine-learning algorithms are in the public domain, training data is not: Most of the voice data used by large corporations is not available to the majority of people, expensive to obtain or simply non-existent for languages not globally spread. The innovative potential of this technology is widely untapped. With providing open datasets, we take away the onerous tasks of collecting and annotating data, which eventually reduces the main barriers to voice-based technologies and makes front-runner innovations accessible to more entrepreneurs.

Having access to voice data opens up a space for an infinite number of new applications. Voice recognition is well suited to reach people living in oral cultures and those who do not master a widespread language such as English and French. With voice interaction, we can reach millions of people and make tech more inclusive.

GIZ is currently exploring voice interaction and mechanisms to collect voice data in local languages. In a first step, we are focusing on Rwanda and Kinyarwanda as the local language. At the same time, English and French is widely spoken in Rwanda and allows for the use of existing open data sets provided by Mozilla’s Common Voice Project (https://voice.mozilla.org).

With the hackathon, we aim at introducing voice interaction to local software developers and language enthusiasts in Rwanda and build a community around this topic. We would like to develop and prototype mechanisms and incentives to collect voice data. The respective solutions could be both technical, based on the infrastructure and resources provided by Mozilla or exploring psychological motivations and barriers using paper-based prototyping.

You may also like:

All solutions and ideas developed during the hackathon will be published under a CC-SA-BY 4.0 licence.

How can we incentivize open voice data collection (in Kinyarwanda)?

If you are …

● familiar with the basic concepts of machine learning
● a seasoned web-developer
● an experienced UX designer
● an app developer
● a social scientist with an interest in motivational patterns
● someone who is professionally working with Kinyarwanda such as a translator, teacher etc. or
● someone who is working on a digital solution to tackle health-related issues in Rwanda

… come and develop with us the basics for voice interaction in Kinyarwanda!

You will have the possibility to learn more about the potential of voice interaction for social impact, with a special focus on Rwanda and the health sector as well as get first hand guidance on machine learning and network with our experts from Mozilla Foundation.

N.B:
We will select 5 teams
The winning team will get Frw 2,000,000
Other 4 teams will get Frw 500,000 Each
team should be composed by 2 to 4 people

Deadline for application: February 1, 2019

Link to Mozila’s Common Voice project : https://voice.mozilla.org/en
Introduction to Speech and Machine Learning : https://research.mozilla.org/machine-learning/

APPLY NOW

Please Share to