Problem with voices

[Jim Loskot]

Hello, new to these boards.

I have got Loquendo simon voice installed and have regenerated all the voice tags. Unfortunately all the prompts have stayed with the previous voice. eg crystal's voice "now browsing artists, the current artist is" then simon's voice "name of artist etc..   ". I have tried various things and have had a good search of the forum but found nothing. I would really like for all the voices to be the same.

Thanks in advance for any help.
Jim

ps
I've got DMS cartridge and using phatnoise music manager 2.35

[VorTechS]

You'll need to generate new prompts for those static phrases and copy them onto the DMS, as I don't believe that PMM does this for you.

I'm not sure about more recent versions.

For info, I just checked my DMS and found that the static prompts are in the PhtSys\TTS directory.  Mine aren't even Crystal - just the original ones by S Crosby.

If there turns out to be no other short term way of doing this, I'll add a DMS Tool into PhatHack Media Manager to do this.

[VorTechS]

Well, had little else better to do today, so I've built a tool to re-generate the static voice prompts in the PhtSYS\TTS directory, where the tool can sensibly do so.

This basically involves some basic simple logic to convert files that are MP3 and are:

Called 'noaac.wav'
Called 'br<number>'
Called 'cur<number>'
Called '<something>.alpha'
Called '<number>'

Do not have '_' in the filename.

Right now I have it generating 48khz 16-bit Stereo text-to-speech WAV files, and then transcoding them to 192kbps MP3 files. 

I notice from the originals, that they are 44khz, 128kbps so will be interesting to see if the new ones work. 

I can easily drop the sample frequency down if not.

Screenshot of the final result:

It also makes a full backup of the existing files before starting the process.

I'll unleash this as part of the next release of PhatHack Media Manager [it will be a bit of a nightmare to separate it out as a stand-alone tool].

I have no idea yet if the resultant files will work with the DMS and frankly don't fancy trashing my DMS until I have my second DMS up and running.

If anyone fancies testing out the resultant MP3's give me a shout and I'll chuck a couple of them over for you to copy onto the DMS.

[Jim Loskot]

Excellent. Thanks for this. I'll have no time for a few weeks but when I do I'll check this out.
Thanks again

[todd1010]

It should work because I've been using the Audrey voices that I created for a long time now. I did this over 1.5 years ago and it works fine without any problems. I just deleted the static phatbox prompts and added the new Audrey voices.

[VorTechS]

Okay, with the lack of volunteers [shame on you!] I tried this out myself.

The Kenwood Keg does not support 48k MP3 files, instead of 'The current artists is' etc, all I got was one of the beep.wav files.

So the next test is going to be:

1) 48k voice announcement file (.wav)
2) Transcoded to 44k audio (MP3)

As it's my turn to drive to the office this week, I'll be trying it out later and will post the results tomorrow.

If anyone has a PhatBox and would be willing to try both of these tests out, please let me know and I'll send you the replacement files.

[Jim Loskot]

not been back to look at this for a while. I'll try and test my phatbox out next week

[VorTechS]

If you were planning to take a look at PhatHack Media Manager to do the work, check out this:

http://forum.phathack.com/index.php/topic,1131.0.html

You might also want to take a look here:

http://forum.phathack.com/index.php/topic,1123.0.html