A couple of weeks ago, I was given the opportunity to do a product review on Dragon Professional Individual version 15. I like to try to go to the product/software as much as I can and test various aspects of its functionality. Dragon Professional Individual focuses on professionals who use speech recognition as a way to increase their productivity.
One of the ways that professionals try to increase their productivity is by using voice recorders. They take notes by voice, it’s recorded to an audio device, using an MP3 file; and then there able to take that MP3 file and dump it into Dragon Professional Individual and have it transcribed.
As I was testing this function, I found a problem.
The supporting documentation for Dragon Professional Individual says that the supported audio file formats are .wav, .mp3, .WMA, .DSS, .DS2 and .M4A. All of the audio formats have to have a minimum sampling rate of 16 kHz.
I created a sample audio file using a text to speech engine [Balabolka]. The output of that application created a 16 kHz audio file [in English].
I tried to process the file with Dragon Professional Individual version 15, but it through an error that the audio file that I was trying to process had to be at least 16 kHz.
I had already reached out to Nuance support about this problem. Thinking that the error might be caused by the text to speech engine that was used from Balabolka, I decided to record my own voice using my WebCam to an audio file/MP3.
I verified the audio file was at 16 kHz, and then I tried to process it through Dragon Professional Individual.
Nuance needs to add the drag-and-drop ability to their DPI software. I was surprised that this didn’t do this.
By the way, there’s no progress bar. You have no idea when this process is going to end.
I got the same error as before.
At this point, I’ve tested two different files that were 16 kHz and neither worked.
I did another test for an audio file at 22050 Hz.
That file failed.
I repeated the process all the way through 48,000 kHz. None of the files that I tested worked.
In continuing my testing, I used the voice recorder and Windows 10 to record a sampling of my voice reading a release of Dragon Professional Individual for the Mac. It recorded my sample file to an MP4a audio file.
I ran that file through Dragon Professional Individual version 15 and it came out fine. For some reason, it processed that file.
The results are interesting, because according to the technical information from Nuance on Dragon Professional Individual’s ability to transcribe audio files, they have to be at least 16 kHz. In my testing, the files I presented that were at or above 16 kHz and they didn’t work.
Apparently, there something more to the types of audio files that Dragon Professional Individual will handle. I don’t know specifically what that is, but I find it interesting that there are more caveats to being able to use any audio file.
If you have any thoughts or opinions on this, let me know in the comments below…