As the Lead Product Designer at Rogervoice, my goal is to maximise usability for our app. The experience of having phone calls by reading live transcription is critical for deaf people.
Therefore the call view is one of the most important.
Our text-to-speech system had proven to be effective, but the overall experience wasn't quite on point yet.
Here are some of the main problems I identified before conducting this redesign.
We removed the background gradient of the bubbles and introduced better hierarchy with carefully chosen typography.
The speaker's bubbles (left) are very important for the deaf person to read, whereas his own bubbles (right) are there for spell checking purposes only.
Uncluttering the interface and clarifying the controls
Most deaf people aren't mute, so the text-to-speech input is a secondary option for our users. I still wanted to clearly provide the option, but the emphasis needed to be brought down. That's why I created the bottom right button.
The same goes for the secondary controls at the top, not used by most of our users. They were regrouped inside a "+" sub-menu.
The gained space is used to add information : the remaining credit (colored blue at the top center) and the speaker's sound wave (white wave at the bottom left-hand corner).
As always, simplicity is the key.
Typography and hierarchy make a huge difference.
Three months after the release : the average call rating went from 3.9 to 4.2 / 5. It worked