Speaksee is a kit of wearable bluetooth microphones that uses speech recognition technology to make conversation visual. For those who have profound hearing loss, Speaksee dramatically enhances speech comprehension by providing subtitles to group conversations in real time.
The fundamental challenge of effective voice to text transcription is dealing with background noise. This is why speech recognition apps commonly struggle in noisy environments or when multiple people are talking.
Speaksee solves this problem through the use of wearable microphones that only hear what the individual is saying. Multiple people can talk at the same time, and their microphones will transcribe what they say, separating their words into different coloured chat bubbles in the Speaksee app.
With transcription accuracy of up to 90%, the result is a low investment alternative to cochlear implants, lip reading or sign language, a new way for those who are hard of hearing to easily follow group conversations.
Because chat logs can be saved, reviewed and searched for later reference, Speaksee applies to a range of other markets as well, from providing trackable corporate meeting minutes, to documentary interview transcription, real time university lecture notes and more.
For a detailed and exhaustive process journal for this project, click here.
© Joshua Flowers, 2015