RadioBOSS 6.3 [beta]

[djsoft]

It happens even with tracks that are not in the database. (I use MySQL).
I will try to find the cause but it always happens, also it is always in the second 0.0

When a track is played, or added to the playlist, an entry for this is added to internal database, with default values, and Mix Start will be set to -1. I did a quick test and this is what happens, in this case Track Tool shows Mix Start as not set. I'm not sure where the "0" value can come from.

Technically it is not possible or very complex to analyze dtmf in the main player without passing it through a sound card?

It's not complex, the problem here is that DTMF analyzes stream as fast as possible, while playback works at its own speed, so it doesn't really solve the sync issue completely.

In windows 11 this does not happen, at least on my test pc, I even tried restarting the computer. (I didn't have time to test the xml parameters yet)

Indeed, looks like a bug. It doesn't use the default voice, it uses the first voice in the list regardless. Now RadioBOSS has voice selection, and this should be solved.

It is indifferent.
If there is a marker visible, and then tracks are moved and it fails to complete the next hour (because there are no tracks reaching the full length of the hour), instead of hiding what would be the right thing to do, start reporting an hour not complete
still happens in latest beta

Thank you, will check.

 

[djsoft]

Now if it does, but I have a question. Why are only the voices that say "desktop" available?

The list is retrieved using Speech API, without any filters, meaning "get all available voices" - so the list is what the system returns. I also see that some voices are missing (5 in windows control panel vs 3 in RadioBOSS), but I don't see any reliable solution for this for now.

 

[djsoft]

Am noticing problems with RB loosing the Main Sound Card we moved to Bluetooth Speakers and this has brought connection issues under Windows 11 It sees Windows said it was connected and RB could not locate it I reset the speakers and it came back very strange !

When you disconnect BT sound card, it's removed from the devices list, when you connect it appears again, and could have a different Device ID. Using Bluetooth is not very reliable. The only way I could recommend is to use "DirectSound" option in RadioBOSS and select "Default" device, in this case it will simply play on default device in the system, following any device changes.

 

[Sasa Grbic]

Greetings to all community members.
1. I want to clarify the situation about TTS voices in Windows 11. I guess RadioBoss only supports sapi5 voices. Windows 11 introduced Neural" voices that sound like natural human speech. However, those voices are only available for Narrator. None of the screen readers can use those voices, so RadioBoss can't use those new Neural voices either.
2. In my previous post, I wrote that I cannot select a task in the scheduler when there is only one on the list. As soon as there are more of them in the list, they can be selected with the arrow. Tonight I was creating a new task and moving through the tab I came to an item where NVDA told me "Time radio button checked". I never paid attention to this option and said let's see what else there is to choose. However, this option cannot be changed with the arrows. Is there another radio button to select or not? This doesn't work on the official 6.2 version either.

 

[nelson c]

When a track is played, or added to the playlist, an entry for this is added to internal database, with default values, and Mix Start will be set to -1. I did a quick test and this is what happens, in this case Track Tool shows Mix Start as not set. I'm not sure where the "0" value can come from.

I did an inspection to the database and these alternate between 0 and null.
In both cases the checkbox in tracktool is shown active

 

[nelson c]

Update: null is replaced by zero on playback

 

[nelson c]

Greetings to all community members.
1. I want to clarify the situation about TTS voices in Windows 11. I guess RadioBoss only supports sapi5 voices. Windows 11 introduced Neural" voices that sound like natural human speech. However, those voices are only available for Narrator. None of the screen readers can use those voices, so RadioBoss can't use those new Neural voices either.
2. In my previous post, I wrote that I cannot select a task in the scheduler when there is only one on the list. As soon as there are more of them in the list, they can be selected with the arrow. Tonight I was creating a new task and moving through the tab I came to an item where NVDA told me "Time radio button checked". I never paid attention to this option and said let's see what else there is to choose. However, this option cannot be changed with the arrows. Is there another radio button to select or not? This doesn't work on the official 6.2 version either.

Azure AI Speech | Microsoft Azure

Explore AI Speech from Microsoft Azure that include speech recognition, text to speech, speech translation, voice-enabled app features, and more.

azure.microsoft.com

These voices are far superior to the narrator voices, they are available in climpcham (pre-installed on windows 11), but not for the narrator.
The improvement is huge, there are python codes that can generate an mp3 with this. I hope that in time this can be used from the api that RadioBOSS uses.
For now it is far superior to use this external tool..

 

[HMMMM]

Well it appears this sound processing is a problem:

https://gyazo.com/d36e09067d3ef9b35b652234a73c2e74

 

[HMMMM]

I doubt it's the highest priority. It would be certainly an improvement, and it's already scheduled for the next update.

Yes, I understood what you said but you also said you had added some other sound features which won't be able to be used as they will not be able to be removed from the path of the DSP. I think these additions are great but surely sound quality is your main goal for you. I am struggling to understand why it's not a issue. for others too? Surely this just needs another port opened to pipe the sound in the direction of the output like the others. Even if you didn't want to add output for Auto Intro could they not share the same output resource as either Jingles or the Cart?

 

[djsoft]

When changing the music library style from "Aqua Light Slate" to "Windows" the following message appears (see attached picture)

Thank you. This is a bug in the styling engine, we'll wait until it's fixed.

1. I want to clarify the situation about TTS voices in Windows 11. I guess RadioBoss only supports sapi5 voices. Windows 11 introduced Neural" voices that sound like natural human speech. However, those voices are only available for Narrator. None of the screen readers can use those voices, so RadioBoss can't use those new Neural voices either.

You can see the voices RadioBOSS can use in the "Add text to speech" window, in the voices dropdown.

Tonight I was creating a new task and moving through the tab I came to an item where NVDA told me "Time radio button checked". I never paid attention to this option and said let's see what else there is to choose. However, this option cannot be changed with the arrows.

"Time" radio button works in pair with "Hours" option - you choose either of those, depending on how you want to start an event. I did a quick test and Narrator reads those options properly.

I did an inspection to the database and these alternate between 0 and null.
In both cases the checkbox in tracktool is shown active

Thank you, this looks like a bug when updating db records for existing tracks.

 

[djsoft]

These voices are far superior to the narrator voices, they are available in climpcham (pre-installed on windows 11), but not for the narrator.
The improvement is huge, there are python codes that can generate an mp3 with this. I hope that in time this can be used from the api that RadioBOSS uses.
For now it is far superior to use this external tool..

I suppose it makes a call to the external server, that is, it's not working locally?

Well it appears this sound processing is a problem:

It means the device that was selected for "Main" sound card was not found, either it's not available for some reason (e.g. disconnected) or its ID changed (because of Windows or driver update).

Even if you didn't want to add output for Auto Intro could they not share the same output resource as either Jingles or the Cart?

As I said, sound routing will be improved in future updates, allowing more control over what plays where.

 

[nelson c]

Thank you, this looks like a bug when updating db records for existing tracks.

Note that tracks with the null attribute are also checked in the checkboxes in tracktool

 

[Sasa Grbic]

Okay... Maybe I wasn't clear. NVDA correctly reads the checked radio button, but when I try to use the arrows to switch to another option, I can't switch. As if there is no other option. It would be logical if there were other or more options because the radio button would not be checked.

 

[Sasa Grbic]

Here is the new information. Now, with the help of the Jaws cursor, I searched the window for adding a task to the scheduler. I saw that the second option was "hours". When I activated this option by simulating the left click of the mouse, I got one unnamed button. With the up arrow I successfully went back to the time option, but pressing the down arrow again failed to switch to the hours option. I need to see in the user manual what can be done with the hours option.

 

[Sasa Grbic]

I suppose it makes a call to the external server, that is, it's not working locally?

Yes. Works online. A fast and stable connection is a must. This works by paying for a certain number of readings per year. You get an api key that is inserted into the application and that's how it works somehow. I don't know the details now. Balabolka text processing and conversion software has this capability built in. You buy a service from Microsoft, enter your api key and use voices.

 

[djsoft]

Note that tracks with the null attribute are also checked in the checkboxes in tracktool

Yes, there's a bug - will be fixed in the next update.

Okay... Maybe I wasn't clear. NVDA correctly reads the checked radio button, but when I try to use the arrows to switch to another option, I can't switch. As if there is no other option. It would be logical if there were other or more options because the radio button would not be checked.

This is because of tab orders, if we make arrows work, then tabbing from "Time" radiobutton will lead to "Hours" radiobutton, instead of "Time" edit box where you set the start time.

I need to see in the user manual what can be done with the hours option.

When you select "Hours" the "..." button becomes enabled where you can set hours/minutes/seconds when the event will run.

Yes. Works online. A fast and stable connection is a must. This works by paying for a certain number of readings per year. You get an api key that is inserted into the application and that's how it works somehow. I don't know the details now. Balabolka text processing and conversion software has this capability built in. You buy a service from Microsoft, enter your api key and use voices.

Interesting. Maybe we'll add this as another TTS option in the future.

 

[Sasa Grbic]

I found a way to activate the hours option with NVDA as well. You need to press the nvda key and - on the numpad for NVDA to say "Move to focus". Then hold the nvda key and with the number 4 on the numeric keypad, move to the hours radio button checked. You still need to hold the nvda key and press * on the numpad to make NVDA say "Move navigation object to mouse". Then you need to press / on the numeric keypad as a simulation of a left mouse click. We have thus activated the hours radio button checked. NVDA key is zero on the numeric keypad. With the up arrow, we can easily return to the time radio button check. It would be better if one of these two options could be activated simply with the arrow up, down or left and right. But when he can't do it like that, it's good that he can do it like this.

 

[HMMMM]

Have written to StereoTool by email to ask them if they can add an adapter port input like VB-Audio Virtual Cable mixing it with Mic in the BIMP as RB wont do this and its using up an input in Windows that is required for the Microphone .

Am still hopeful that Dimitri will add Auto Intro to this group of Play settings to have a shared adapter port with VB-Audio Virtual Cable that DSP could read without going through input on Microphone. Currently this only does this with Jingles and Cart

Wall sadly I don't understand what is difficult in adding it as the code must be the same for Jingles and Cart Wall so must already exist perhaps Dimitri could explain what makes these Auto Intro so much harder so we could understand as it seems simple moving of existing tested code to us and the placement of an extra box on Settings>>Play? If he would like to send code, I might be able to do it Cheeky grin

 

[Stefan]

Hello, Maybe it is mentioned before but i noticed that "process music tracks'' under 'Tools' in Music Library doesn't process the selected tracks but the whole music library. In previous versions of Radioboss the selected tracks were processed not everything in the database.

 

[nelson c]

Hello, Maybe it is mentioned before but i noticed that "process music tracks'' under 'Tools' in Music Library doesn't process the selected tracks but the whole music library. In previous versions of Radioboss the selected tracks were processed not everything in the database.