Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: PSK31 Using Voice to Text  (Read 2837 times)
W7HC
Member

Posts: 3




Ignore
« on: May 11, 2004, 10:17:52 PM »

I would like to know if anyone has used a voice to text program such as Dragon NaturallySpeaking to send PSK31 text. I use Dragon NaturallySpeaking and it works very well for me. It uses a sound card and of course so does PSK31. I only have one built in sound card. Before purchasing a second external sound card, I decided I would inquire if anyone had attempted this before I plunged blindly ahead. Any tips, suggestions, sites, etc. will be greatly appreciated.

Thanks,

Harold-W7HC
Logged
N8IK
Member

Posts: 64


WWW

Ignore
« Reply #1 on: May 12, 2004, 12:27:00 PM »

This sounds very cool - keep us posted!
Logged
KG4HLZ
Member

Posts: 21




Ignore
« Reply #2 on: May 15, 2004, 01:57:04 AM »

I am experimenting with VR in Ham Radio as well. If I understand your idea correctly, you will be using Dictation more. The state of the technology is still quite crude for this. Don't let me discourage you though. The Microsoft Speech SDK is available for free as a download from there site. There are a few for Linux as well.

Let me know how things work out. We may be able to exchange ideas.

73s
Bruce
KG4HLZ
Logged
W7HC
Member

Posts: 3




Ignore
« Reply #3 on: May 15, 2004, 11:29:06 AM »

I'm not certain what VR is. I have a software program, Dragon NaturallySpeaking, that I have been using successfully to covert speech into printed files, emails, etc. I am speaking into a microphone at this moment as this email is appearing on my screen. I have used this voice to text program with MixW. The words appear in the send window just as they are appearing in this email by speaking into a microphone plugged into my computer. And the words then appear in the sent window of MixW just as when it is typed in. But I have no way to connect my radio to my computer as my only sound card is already in use by having the voice to text mic plugged into it. I do not know if it is possible to use two sound cards at the same time. If it is possible, I could buy a second external USB sound card and perhaps send PSK without typing. I was hoping to learn if anyone else had tried this before I spent money on a second sound card. Over............
Logged
N5EIL
Member

Posts: 121




Ignore
« Reply #4 on: May 15, 2004, 01:08:27 PM »

Now here's an idea, use the voice to type for psk-31 and then find or create a program that reads what you typed! Now you have an ultra narrow voice mode!

N5EIL, Neil
Logged
KG4HLZ
Member

Posts: 21




Ignore
« Reply #5 on: May 15, 2004, 02:30:49 PM »

Sorry, VR is voice recognition.

If you are getting a good dictation mode text to speech, then maybe I should look at Dragon Dictate. I have only been able to get about %80 accuracy on a good day.

I know that Windows can support 2 sound cards but I have never personally set it up. I know that the other VR systems that I have used allowed you to select the input device. I am assuming that Dragon will do the same.

Bruce
KG4HLZ
Logged
W7HC
Member

Posts: 3




Ignore
« Reply #6 on: May 15, 2004, 04:06:54 PM »

Exactly Neil! My dragon program will read text back to me. Narrow bandwidth, the low transmit power requirements, and lots and lots of room! And Bruce says Windows will support two sound cards. So I guess I'll just have to buy a Sound Blaster external sound card and give it a try. If it doesn't work, it will not break the bank and the enjoyment derived will probably be worth the cost.

By the way, I have only made three corrections in this post. I'm using a microphone only.

Harold
Logged
KG4HLZ
Member

Posts: 21




Ignore
« Reply #7 on: May 16, 2004, 04:40:02 PM »

Thats a great idea!!

Since you are only sending text to be re-synthesized into voice on the other end. I wander if you could get better results by using a phonetic spelling instead of speech to English spelling. If the system miss recognizes something, then it will get something that sounds close at least. Just a thought. I don't even know if phonetic spelling systems are out there.

Bruce
KG4HLZ
Logged
KB7YOU
Member

Posts: 42


WWW

Ignore
« Reply #8 on: May 24, 2004, 08:49:28 AM »

Hi,

You can use multiple sound cards in a Windows box. I do it frequently for a variety of reasons. To keep things easy to setup be sure to use sound cards that are very Windows compatable. In other words, if you have to install special software or non-certified drivers then avoid the card.

I use multiple Sound Blaster cards including an Audigy and an old SB16 card without problems.

Just about any properly written Windows program will ask you wich card you want to use. If the program does not do this then Windows lets you select a default card for output and input.

Chris
KB7YOU
Logged
N8NXG
Member

Posts: 1




Ignore
« Reply #9 on: December 26, 2005, 07:08:01 AM »

I know this reply is to a very old post. It's the day after christmas and I don't have to work so I was just reading all the posts.
I had dragon naturally speaking set up to run with digipan. The OS is XP pro. I had two sound cards installed. One was an aureal and the other was a soundblaster. Windows has no problems with this set up. I installed the drivers for the aureal card and I let windows install it's own drivers for the soundblaster. As I recall dragon would let you select what card you wanted to use. Digipan at that time only used the default card. The new version of digipan (2.0) will also let you select a card.
I started dragon first and then minimized it. I then started digipan and clicked in the send frame. As I recall I had to put digipan into transmit and receive mode. The mic I used had an on-off switch. If you have one use it. Dragon will fill the transmit frame up with whatever it hears even when digipan is in receive mode. It worked very well. If I had to guess I am a day late and a dollar short with this reply but who knows

Steve  
Logged
WA6NUT
Member

Posts: 8




Ignore
« Reply #10 on: October 10, 2008, 11:12:20 AM »

This too is a reply to an old post.

An excellent article on using Speech Recognition in ham radio appears in the April 2006 issue of QST, pp. 55-56.  Written by Jim Gluckin, WA2LPF, it describes the use of DigiPan or MixW with DNS speech recognition software.  Of particular interest is the use of a USB headset/mic so a second soundcard isn't required.

For text-to-speech (TTS), a version of YGPSK has been developed by N7YG for use with Microsoft SAPI 5.1 and later (I've also tried DigiTalk by KH6TY, but I can't get it to work with Windows XP).

I've been using MixW 2.18 for TX/RX together with YGPSK 4.6.4.5 for RX only.  For speech recognition/TTS I use the Microsoft Speech Recognizer 6.0 (also known as the Speech Add-On for Internet Explorer).  I use a GigaWare USB Headset/Mic (and a Cyber Acoustics USB speaker for listening to PSK31 from another room).  A fairly fast PC is required (I use a 2.4 GHz Compaq desktop).  I've posted details of this setup on my HamSDR Download page at:

http://www.hamsdr.com/PersonalDirectory.aspx?did=1139

Listen for me on 20M VoPSK (Voice over PSK31)!

73, Rick, WA6NUT



Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!