Voice notes with text navigation (idea #96)

Many phones incorporate a "voice memo" feature that allows digital voice recordings to be made. However, often this feature is next to useless. To start with, open up the recordings folder and the user is often confronted by a long list of obscure file names, like 24200011.amr, 24200012.amr and so on. It's like the old days of computing all over again, before computer scientists realised that normal people were trying to use their software and before usability - "design" - became vogue (obvious?).

The key missing ingredient is voice recognition. This is not beyond the realms of a mobile phone's processing capabilities, especially with today's powerful low-cost processors. Some phones today even have graphics accelerator chips in them to make 3D games go fast!

What I think is worth exploring on the mobile is a hybrid approach to allow speech files to be navigated using a text-driven interface, but reviewed and edited using voice. For example, I record a series of voice notes and can then look through the notes using the first few words from each note. I imagine being able to spin through the opening words of each note using a thumbwheel and then clicking on the entry to hear the voice note again. If the voice note is large, I can use a similar tool to navigate through the note using "text clips" of the converted voice stream.

Using this approach, the use of voice as a tool for note taking becomes viable. It also means that it is not necessary for entire audio files to be converted, which might be too costly in processing terms, especially if fast and accurate performance is required. Performance and accuracy can be improved by uploading the voice files to the PC for some serious number-crunching offline.

An additional feature to be considered is voice search, using a spoken word to find matches in the voice notes.

Buy my book (Amazon US/UK)
Join my email list
Subscribe to my "100 Mobile Product Ideas" free e-book

Tags: mobile, voice, speech recognition, note taking, memo