iDictate – Digital Voice Blog

No, is the short answer. Voice or speech recognition software will only successfully work with one spoken voice and that spoken voice, for best results, needs to have trained the software to understand their voice and style of speaking known as creating a voice profile. Most people who use voice recognition software typically pop on a headset and mic and chat away to their Windows PC or Mac and watch as their voice is magically (or using the Dragon voice recognition engine if you don’t believe in magic) converted to text.

Voice recognition is currently going through a spike in popularity, especially since the release of the iPhone 4S running iOS5 with Siri, which has raised awareness of just how good voice recognition technology has evolved over the last couple of years, Nuance being the company leading the way in this technology on Mac, Windows and now inside iOS5 working closely with Apple.

Because of this everyone now has an iPhone which is capable of recording audio and most who need to capture important meetings or interviews also have a high quality digital voice recorder. Dragon NaturallySpeaking Premium and Professional edition on Windows and MacSpeech Scribe on the Mac can take recorded audio and transcribe it, that is only for that single trained person who is speaking. So it is understandable that the leap is often made that if it can transcribe one voice it can transcribe many, not so. The technology is not there yet and in my opinion won’t be for quite some time.

More

With all the flurry of activity over the last week with Mac OS X 10.7 (Lion) being released along with an upgrade to Dragon Dictate from 2.0 to 2.5 we have had a few calls into our office regarding compatibility of MacSpeech Scribe with Lion.

Nuance have recently posted an update on their website (click here) which is important to read if you are one of these customers:

• Dragon Dictate 2.0 – upgrade to Lion YES
• MacSpeech Dictate 1.5 – upgrade to Lion YES
• MacSpeech Scribe – upgrade to Lion NO

Upgrading to Lion on your Mac should be fine if you are running Dragon Dictate 2.0, I have personally done this and my Dragon Dictate is working without issue. Updating to Lion if you have MacSpeech Dictate 1.5 Nuance also claims should be fine, I have not tried this.

But, if you are a MacSpeech Scribe customer the advice is to not upgrade to Lion from 10.6 (Snow Leopard).

MacSpeech Scribe is the legacy MacSpeech product now maintained by Nuance which allows voice to text from recorded audio only. Personally I am surprised that with Dragon Dictate 2.5 the ability to transcribe from spoken voice AND recorded voice was not included, I am sure this has to have been high on the Nuance Mac teams priority list for some time but as yet is still not there.

Keep an eye on my blog and the official Nuance Dragon for Mac website for any updates.

Beautiful Lion Tamer picture (without the MacSpeech Scribe box) oil painting by Jennifer Li, click here to see Jennifer’s amazing artwork.

Nuance has quickly forged their authority on the Dictate brand of voice recognition software for Mac and with their first release Dragon Dictate 2.0 they have done an excellent job. But what now for the Legal and Medical edition still stuck at MacSpeech Dictate version 1.5 and the fairly new offshoot product MacSpeech Scribe currently at version 1.1, all of which are using the older Dragon speech recognition engine and not the new one included with Dragon Dictate.

I have some theories for these products which may or may not be correct so please don’t hold me to them. These are my best guesses and have not been confirmed or denied by Nuance, just my two cents worth.

MacSpeech Scribe in my view can go in ne of two directions. It can either be discontinued and the voice to text functionality built into a later Dragon Dictate update, much like Dragon NaturallySpeaking Premium or Professional for Windows. Or Nuance will embrace this as a separate product and continue to push it as a stand alone offering for transcribing only recorded audio. I have always been torn on MacSpeech Scribe, I have never understood why you would create a separate product for transcription of digitally recorded audio but that maybe because I have been so used to the functionality from the Windows side of the fence. As MacSpeech Scribe is still fairly new I have a feeling Nuance will continue with it and we will likely see a Dragon Scribe product appear in the very near future.

So what about the Medical and Legal version of MacSpeech Dictate, are we going to see an equivalent in Dragon Dictate? My personal view is that we are likely to see a Legal version for Dragon Dictate but maybe not the Medical version. From my understanding the Legal dictionary of terms is fairly standard and so can be quite easily built into a speech recognition dictionary. Wheres Medical with its large number of specialised area of understanding would be an unwieldily dictionary to maintain and build. For the Medical side of things I think we will see a third party put their hand up and claim an exclusive partnership with Nuance to provide and support that medical crowd.

So when can we see what I have theorised about above. Dragon Scribe I would think would be very close, if it is coming, and we would likely see this before the end of the year. Dragon Dictate Legal I would also see on the same time line, especially as we now have Dragon Dictate in the wild adding a more comprehensive legal dictionary can’t be that far behind.

Of course as soon as we get wind of any new voice recognition products from Nuance for the Mac or Windows we will post the info.

MacSpeech Scribe just released by MacSpeech lets Mac users record their voice to a digital voice recorder and then have MacSpeech Scribe transcribe it for them. The only certified digital recording devices for MacSpeech Scribe are the iPhone, iPod Touch (2G+ plus a mic) and some of the Olympus digital voice recorders.

Two of the Olympus digital voice recorders certified for use with MacSpeech Scribe are the Olympus DS-3400 and DS-5000, both from the Professional series of digital dictaphones from Olympus, these are serious digital dictaphones. They both come with DSS Player Version 7 Plus for Mac which simplifies the task of getting the audio off the recorder. The software lets you also setup users on the recorder, set noise cancelation levels, playback and edit your recorded dictation and direct record using your digital dictaphone as a high quality mic.

Out of the box both the DS-3400 and DS-5000 will record in DSS Pro digital audio format, these files have a suffix of .ds2 – files of this format can not be loaded into MacSpeech Scribe so some file conversion needs to take place, this is where DSS Player Version 7 Plus can help. On a Mac it is not possible to set the DS-3400/DS-5000 to record in .aif format by default.

For file conversion you have two options. 1. you can manually convert each file 2. you can set DSS Player Plus to automatically convert audio every time your DS-3400/DS-5000 is connected. Both are very simple:

Manually Convert DS-3400/DS-5000 Audio To .aif Format

To manually convert .ds2 audio files to .aif on a Mac using DSS Player Version 7 Plus simply right-click on the .ds2 file you want to convert and select Convert to AIFF File, your converted file will be placed in the same download folder tray as the original file. If you want to save the .aif file to an alternative location simply highlight the audio file to be converted and click the Convert to AIFF File icon .

Automatically Convert DS-3400/DS-5000 Audio When The Recorder Is Docked

If you prefer to convert every audio file during download to your Mac when your digital dictaphone is connected then open the DSS Player Preferences and click the Download tab. Here you can tick “Convert to AIFF file during download” which will have the desired affect. Your .aif files will appear in the download tray folder configured and are then ready to be loaded into MacSpeech Scribe.

At the recent MacWorld 2010 Expo in San Francisco, the place for companies developing technologies for the Apple Mac and iPhone to show off their ideas, MacSpeech announced the launch of their new software product MacSpeech Scribe.

MacSpeech Scribe will do what alot of existing MacSpeech Dictate customers have been asking for, transcribe from a digital recording of their voice. Now you no longer have to be in front of your MacBook or iMac to dictate audio to be transcribed. Using one of the certified MacSpeech Scribe digital voice recorders you can record your thoughts anywhere. Then at a time convenient to you have MacSpeech Scribe transcribe the audio for you.

This is similar concept to the Philips LFH0667 (formerly LFH0660/10) which was a cut down version of Dragon NaturallySpeaking 10 for Windows which would only take audio from a digital recording.

MacSpeech Scribe is a stand alone product and is not an add-on to MacSpeech Dictate. If you already own MacSpeech Dictate and have created and trained a profile for your voice that profile can not be imported into MacSpeech Scribe, you will have to start from scratch with a new MacSpeech Scribe profile.

Only a small number of digital voice recorders have been certified by MacSpeech to work with MacSpeech Scribe. These are the iPhone, iPod Touch (2nd Gen up with a mic) and a few Olympus digital voice recorders. Of the Olympus recorders on the certified list only the DS-3400 and DS-5000 are currently available in Australia. MacSpeech Scribe can take audio in .aif, .aiff, .mp4, .m4a, .m4v and .wav format so in theory a huge number of other voice recorders will be compatible (although not certified). Remember, its all about the audio quality, so just because the $20 USB recorder you bought off eBay records in one of the supported audio formats the quality of that recording may leave you disappointed – stick with Olympus and you won’t go wrong.

MacSpeech Scribe is available now in Australia for RRP $199 incl. GST from Dictate Australia.