I haven't tried it (even my computers are tired of listening to me talk). But, since it generally works as a keyboard replacement, it is likely to work with Sesame much as it would any other software.

Three comments:

First, Vista and Windows 7 both include speech recognition that supposedly is fairly good.  The reason they aren't mentioned is that when Vista was first introduced Bill Gates tried doing a live demo at a major conference and it cratered.

Second, I have had Dragon versions 9, 10, and 11.5 (the latest). I can't use them at all although they are quite accurate.  (Frankly, I'm amazed at the technology.  I spent a year in a doctoral program in linguistics--the science of how language works--so I know how complicated it is.  For both language generation and language comprehension the human mind basically has an incredibly complex computer program running on amazingly complex low-power, high-performance hardware.) I'm a lawyer, and the problem I find is that my writing style changes completely when using dictation technology.  My style when typing tends to use compact short declarative sentences.  When dictating, it comes out as standard legalese.  I spend so much time having to think about the exact wording and then changing it after dictating that it's actually faster just to type.

Third, I recently ran across a free Windows macro program called AutoHotkey.  http://www.Autohotkey.com I've only used it a little bit but from looking through the Help it seems to be incredibly powerful. It could probably be used to automate moving text between programs, as well as many other tasks.

I just tried it and I'm getting "sorta works" results. I have a dual-core Pentium E5700 with 4 GB DDR3, Asus P5G41T-M LX (LGA775) mobo, a 60GB SSD for the OS and a 1.5TB Seagate hard disk for data, PCIe video card, Win 7 Home Premium 32-bit and Dragon 11.5 Home. Although it is recognizing speech in text fields, it's only picking up some of the words. For instance, I spoke, "This is a test to see if Dragon can pick up my speech at a reasonable rate" and it typed "this is  to see if  at a reason" (although I spoke slightly slower than normal, which is not normally required when using Dragon).  The entire sentence comes out fine dictating directly into FireFox here with Sesame running.

I'm not using any memory or CPU-intensive programs (except Dragon).  The hardware and software I'm running is a bit souped up compared to a recent office PC.

I just tried all of these programs in the order listed, each opened successively without closing the others, and speaking at normal speed and each took the speech accurately:

1. FireFox 8
2. WordPerfect X4
3. Word 2007
4. DavkaWriter 7 (a multi-lingual wordprocessor)
5. Acrobat Standard 9 (adding text box)
6. Nuance PDF Converter Pro 6 (Acrobat clone, adding a text box)

Obviously, the problem is the speed of the Dragon-Sesame text interface.

The following does work fine:

1. While in Sesame say "Start Notepad." (Notepad starts.)
2. Dictate into Notepad.
3. Say "Select all.", "Copy that.", "Close.", "Don't Save."
4. Switch to Sesame to the text field.
5. Say "Paste that."

You can say "new line" before paste that, which does work okay.

With Dragon activated and Sesame in a text field, say, "Press F six."  The editor opens.  Dictate your text, then say, "Close."  The text is transferred.

Although the "press" and "close" command work with Windows 7 speech recognition, it won't take dictation in the window.

By the way, Amazon has Dragon ver 10 for $30.  I have versions 9.5, 10 and 11.5 and I don't see a huge difference.  Dragon does come with about 6 dialects, including Midwestern American (standard newscaster), Southern American, UK, Australian, India and maybe 1-2 others.  (I think Singapore too, not sure.)

question - if I speak "1235" will the Dragon software input numerals or words ("1;2;3;5", or "one;two;three;five") ?

I'm installing Dragon on a new computer.  It has the following English dialects for voice recognition:

1. Southeast Asian
2. U.K.
3. U.S.
4. Indian

It has the following dialects for text-to-speech:

1. U.K.
2. U.S.

Something that would be interesting to try and should work would be a program code button that does the following:

1. Set the clipboard to the text in a field.
2. Call a text-to-speech reader like TextAloud and have it set to read aloud the contents of the clipboard.
3. Return control to Sesame

Rick_R wrote on Jan 7^th, 2012 at 7:46am:


Clarification regarding Dragon 10 (still available from various sources).

Dragon 10.0 and 10.01 don't work properly on any version of Windows 7 or 64-bit versions of Windows. When you go to the Nuance site to check for upgrades it takes you to a form that requires your 10.0 serial number.  It then emails you a link for a single download of Dragon 10.1, which is compatible. Only one download per serial number is available.

I downloaded the Preferred version (full retail $200).  Regarding voice recognition, it only has the U.K. and U.S. dialects, not also Southeast Asian and Indian.