Second thoughts on voice-to-text blogging from my iPod Touch with Dragon Dictation — and my Wish List for an upgrade

Update Aug 19

This morning, Nuance announced the release of an upgrade, Dragon Dictation 2.1, here. It is being announced as the restoration of the app’s VoiceOver compatibility (Apple’s screen access technology for the visually impaired — a keyboard driven interface with speech accompaniment), which was degraded in the 2.0 upgrade. So I don’t expect any improvements to the  voice-driven editing interface in this upgrade along the lines of what I discuss below, but . . . well, maybe I have just a teensy hope. I’m off to download it in iTunes and take it for a spin on my iPod Touch. Back with a report later . . .

Now that I have used the Dragon Dictation iPhone/iPod app for a couple weeks, the honeymoon is over. I have found its weaknesses and come up with a wish list for an upgrade. I still think Dragon Dictation is the best voice-to-text transcription mobile app — but if you do any editing, it is a w-a-a-a-y too slow a process for blog posts or much of anything beyond a paragraph. But, then again, at least it provides a possible way to keep blogging if, like me, you have a hand injury and would be out of the blogosphere otherwise.

I'm dreaming up my wishlist for a Dragon upgrade

The Dragon “voice-driven” editing interface is in need of an overhaul if it is to become a serious blogger’s tool (see my four-part “Blogging by Voice” series on voice-to-text blogging for from my iPod Touch starting here.) Such an improved interface would make a world of difference not only to bloggers, but to anyone with physical challenges that make keyboard use difficult or impossible.

Nuance Communications said in a blog post last week that it is committed to “researching optimal use of voice correction and text-to-speech for optimal eyes-free use” in the DD app and welcomes input on how to do so. Well, since you asked so nicely, Nuance, here is “Bay’s Wishlist of Improvements to Dragon Dictation 2.0.”

Bay’s Wish List for Dragon Dictation 2.0 upgade

1.  Spoken command recognition. As of now, the only kind of voice-driven editing in DD is replacement or insertion of text in the current dictation at the point where the cursor is placed. This limitation has caused me to restrict my use of DD for first drafts only. The only way I have found to voice-edit a second draft is to re-dictate the draft entirely, or to dictate small inserts which I then cut and paste into the first draft (which sits on my blog as a draft). This is time-consuming enough to be a deal-breaker for all but the most patient of bloggers (count me out of that group — LOL). An interface is needed that allows one to say, for example, “Change X to Y.” Or “Delete X.” And especially “Scratch that.”

2. Ability to voice-edit an existing document. How wonderful it would be to  open my blog post draft (dictated via DD) and edit a second draft by voice. Currently, voice-driven editing is confined to the document you are dictating. And that document disappears once you exit the program or process your dictation. So there is no way to edit that document by voice a second time. And second drafts are indeed necessary due to DD’s rather dismal word recognition, leading to my final wish . . .

3. Improved word recognition. Build in a learning capability, similar to the training program in the desktop versions. For the type of content I dictate, the mobile app’s initial success rate is about 60%, rising to about 75% in a re-dictated version — which makes improved editing capability crucial. I include a humorous example below.

What I dictated:

Hosted blogs are wonderful, and I very much appreciated them as my “starter home,” but you soon get tired of WordPress making all the decisions for you about themes, plug-ins, widgets, and the like.

What Dragon Dictation transcribed:

Posted dogs are wonderful, and I very much appreciated them as my “stagger home,” but you soon get tired of WordPress making all the decisions for you about themes, plug-ins, widgets, and the lake.

Respond to this post