Not sure what you mean, exactly. The professional Nikons (D1,D2,D3,D4 series) have always allowed you to record up to a 60-second voice annotation to images. The files are saved as a .WAV audio file with the name of the image automatically attached. These cameras all have a built-in microphone for in-field playback and confirmation, and can be re-recorded if needed by selecting the file, manually deleting it, then re-recording a new audio file.
It's super-handy when out and about, or when doing experimental stuff or using non-CPU lenses or extension tubes, or unusual set-ups. I use the voice memo feature all the time. WHen doing PJ work, I used to use it to record information about events, like time remaining in games, who scored, who was thrown out, who finished in what order, ID's for captions, and so on.
If you mean an app to record metadata to photos and to insert it into the EXIF structure...well...not sure if there is something like that based on voice command...maybe there is though.