Re: A Synthesized DVK for N1MM+ - Version Release

Matt Murphy

I used it in NAQP SSB as well and it works quite nicely.  I'm looking foward to playing around with more of the settings.  Very nice work. 

Matt NQ6N

On Sun, Jan 21, 2018 at 7:53 PM, Gerry Hull gerry@... [N1MMLoggerplus] <N1MMLoggerplus@...> wrote:

New release -- thanks for all the great feedback.  Despite bad prop, used it some for NAQP SSB, and got good audio reports.

73, Gerry W1VE

W1VE Synth DVK for N1MM+ Release Notes
- Fixed bug when trimming SCP callsign files (tnx WC2L for testing)
- Completely refactored WAV file creation after Polly synthesis.  Was not producing the
  correct file.   Now generating 16 or 8 Khz Mono WAV files (tnx K3CT for feedback).
- Added support to create Any WAV file in the op directory, allowing creation for keys
  other than the "standard ones". (tnx WB9SND idea) 
- Added the ability to synthesize just prefixes from SCP.  This is enabled without Amazon Keys.
  Many less files than doing full callsign.
- Fixed bug when you add your own Amazon Keys.  Was not enabling creation from SCP files. (Tnx WC2L for test)
 - Added feature to select High or Low Quality Audio files.
Version Release
 - Uses the Microsoft .Net Framework Version 3.5 (Same as N1MM+) and requires internet access.
 - All audio generated is now 8KHz, 16-bit, generating very small files.
 - Added support for the S&PEXCHANGE file
 - Added support for trimming the beginning and end of created files.  Do not exceed 150mS
   or clipping will occur.
 - Added support for Insering a pause between text.  In the text, add

, and specifiy the pause

   time in the Pause value.
 - Added a checkbox that, when checked, will ask before overwriting any of the files (be careful, a lot
   of questions if creating Letters files.
 - Added full support for the Amazon Polly suite of Languages..
   24 languages and 47 voices.  There are many male and femail voices.  The best, default voice
   for US ops is English (US) - Matthew.   You can use foreign languages to synthesize English
   with a foreign accent.
   The user interface has been changed to support full unicode for
   non-western languages.
 - Added support for varying the sound of a voice:
  - Added full support for all speeds supported by Poly.
  - Added Timbre support
  - Added Pitch support
  (Play with these controls to get a sound you like.)
 - Added support to create all callsign.wav files from the SuperCheckPartial database of callsigns.
   Amazon limits free access to the Polly synthesizer; for normal contest and Letters
   creation, this is fine.  However, if you want to create your own custom files, you
   can apply and get Amazon Polly credentials; this will give you 5 Million characters
   a month for a year, after that it's $4 per million.   In order to generate custom 
   files from, you need to enter your Amazon Credentials.
   I have created an SCP WAV set, using the English Matthew Voice and the SCP US/VE/Territories
   database.  This file is located at
   It contains some 18,200 wave files, and is 438Mb is size.  Typical callsign files are 50K or so.
  - Added a Phonetics Override editor.  If you don't like the phonetics I've chosen for a letter,
    you can make changes here.  Pick a letter from the drop down, type the text and press Create.
    Add/Remove from the list.

  - ALL Parameters are SAVED to the configuration file, W1VE-SYNDVK.exe.config.  As soon as you
    make a change, it is saved.
  - Fixed Tabbing issues between controls.

I've loved playing with voice synthesis for years.  Amazon has made very high quality synthesis available
for free (for limited use like ours).  

Thanks to N1MM+ and his dev team for critical feedback on the initial versions of this software.
See ya in the piles!  

73 Gerry W1VE

Join to automatically receive all group messages.