steve0622
Member
ok....thanks for the advise...will give those options a try and continue testing.
TwoToneDetect New TwoToneDetect in the works - Python based
- Thread starter aaknitt
- Start date
-
- Tags
- twotonedetect
A few comments. First, I'm not familiar with the Wouxoun radio, but if it outputs the A and B sides onto the same audio line (as opposed to a stereo line that you're splitting and running into two different sound cards), you could run into problems if there is radio traffic on both channels at the same time...this could cause missed tones. For another $57 ($50 for another radio and $7 for a USB sound card), you could have another dedicated receiver channel that would increase system reliability.
Second, I'm not sure why you're getting both MP3 and AMR. Something to try would to add a second non-phone email address to the MP3 list. If you only get the MP3 to that address, then I would suspect that your phone provider may be converting the MP3 to AMR for you and sending it since some phones won't play MP3 files. If you get the MP3 and AMR to your email address, then you've found a program bug that I haven't seen and I'll need to investigate further.
Finally, you don't have any email addresses for your Long Tone. Not sure if you knew that or if it was an oversight.
Andy
Thanks for the clarification. Now I am wondering how your program knows to look a the last 400ms of the A tone. All the tones that i monitor for start with about 500ms of dead air transmission (probably the PL tone?) just prior to the start of the A tone. If the program is calculating where to start looking for the A tone, at what point does it start that calculation, the start of audio or the start of the A tone? In my case those two points are 500ms apart.
steve0622
Member
aaknitt:
On the Wouxoun radio, it is just a temporary setup mainly to discover the correct paging tones for all the county wide departments using Audacity, as there is not much traffic on fire and ems currently. After I capture all the correct tones, and test the TTD here at the house, then the TTD will be moved over to the fire dept. computer with a radio to monitor only the fire frequency and the seperate ems computer where the TTD will be monitering only 1 frequency as well. Running the dual band radio with 1 sound card is really only for my benefit to complete this first phase of testing. I do know there is the possibility of missed calls, which is why this setup is not being "activly" used yet.
On the MP3 and AMR deal, I have checked in my sent folder for gmail, and it is just sending the MP3 to my phone, so, im not really sure whats happening there. I have just added an alternate email address under the MP3 tones that I will check to see what it is recieving on my suddenlink email account. I have also added another phone number to the list with a different carrier and different type of phone to test that way out as well.
On the long tone, that was an oversight that i noticed after i posted the config file which has been corrected now as well.
Thanks guys for all the support...for the Comtekk system the fire dept. has been using since august, I have not been able to reach tech support with them since we purchased the liscense from them......kinda crappy tech support for a high priced system....but...this TTD is turning out to be much better and i'm liking this software alot more than the other.
On the Wouxoun radio, it is just a temporary setup mainly to discover the correct paging tones for all the county wide departments using Audacity, as there is not much traffic on fire and ems currently. After I capture all the correct tones, and test the TTD here at the house, then the TTD will be moved over to the fire dept. computer with a radio to monitor only the fire frequency and the seperate ems computer where the TTD will be monitering only 1 frequency as well. Running the dual band radio with 1 sound card is really only for my benefit to complete this first phase of testing. I do know there is the possibility of missed calls, which is why this setup is not being "activly" used yet.
On the MP3 and AMR deal, I have checked in my sent folder for gmail, and it is just sending the MP3 to my phone, so, im not really sure whats happening there. I have just added an alternate email address under the MP3 tones that I will check to see what it is recieving on my suddenlink email account. I have also added another phone number to the list with a different carrier and different type of phone to test that way out as well.
On the long tone, that was an oversight that i noticed after i posted the config file which has been corrected now as well.
Thanks guys for all the support...for the Comtekk system the fire dept. has been using since august, I have not been able to reach tech support with them since we purchased the liscense from them......kinda crappy tech support for a high priced system....but...this TTD is turning out to be much better and i'm liking this software alot more than the other.
It doesn't care when the A tone starts. It's continuously sampling audio. In the case of the example in the picture, when it sees a sequence of audio that includes .6 seconds of A tone immediately followed by 2 seconds of B tone, it will generate an alert for that tone set. It doesn't care what comes before or after that audio sequence.
Andy
gfdfortynine
Member
Where do you set the A tone length? I'm not seeing that option in the editor.
Last edited:
Gotcha. And the log.txt file lists the freq for each of those 200 ms samples. Set the atonelength to 0.2 and it will look for a btone after detecting the a tone in one sample. Set it to .4 and it will require two a tone samples consecutively before the b tone, etc
For now you'll have to manually add that parameter (Atonelength) to the tones.cfg file. If we determine that the v59 algorithm works well, I'll redo the TonesEditor program to add that parameter and remove the debounce ones.
Andy
Dorf411
Member
I have been using 58a for a couple weeks testing and just today decided to try and switch up to ver 59. A little later there was a tone out for a fire dept that uses a longtone of 433.7 for some reason this caused the program to detect and send a message 4 times. Any thought what may have caused that?
I have been using a tone length of 7 and debounce of 3 when I put it back to 58b everything worked fine again.
I have been using a tone length of 7 and debounce of 3 when I put it back to 58b everything worked fine again.
Interesting. I knew that could be an issue with the long tones but I thought I had a fix implemented that would prevent it. Were you using the ignore_after parameter at all? The program is supposed to automatically set ignore_after to 10 seconds for a long tone to prevent what you're describing. If you can post or send me your tones.cfg file that would be helpful. Also, your log.txt file if possible.
Thanks,
Andy
Dorf411
Member
Tones.cfg
[Tone1]
Atone = 368.5
Btone = 539.0
Btonelength = 3
Btonedebounce = 2
Description = Amb
mp3_Emails = "JohnD"<xxxxxxxxxx@vzwpix.com>
exclude_from = 22:00
exclude_to = 07:00
ignore_after = 60
[Longtone1]
Longtone = 433.7
Longtonelength = 7
Longtonedebounce = 2
Description = Fire
ignore_after = 60
mp3_emails = "JohnD"<xxxxxxxxxx@vzwpix.com>,"Grt Sheriff"<xxxxxxxxxx@vzwpix.com>,"Kevin Chief"<xxxxxxxxxx@vzwpix.com>
Log.txt
uccesfully retrieved local tone info
Succesfully retrieved local tone info
Tone1
Longtone1
defaultdict(<type 'list'>, {'Fire': array([ 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7,
433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7,
433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7,
433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7,
433.7, 433.7, 433.7]), 'Amb': array([ 368.5, 368.5, 539. , 539. , 539. , 539. , 539. , 539. ,
539. , 539. , 539. , 539. , 539. , 539. , 539. , 539. ,
539. ])})
defaultdict(<type 'list'>, {'Fire': 1, 'Amb': 0})
247
516
1064
274
494
962
387
360
360
457
838
387
666
252
424
306
768
[Tone1]
Atone = 368.5
Btone = 539.0
Btonelength = 3
Btonedebounce = 2
Description = Amb
mp3_Emails = "JohnD"<xxxxxxxxxx@vzwpix.com>
exclude_from = 22:00
exclude_to = 07:00
ignore_after = 60
[Longtone1]
Longtone = 433.7
Longtonelength = 7
Longtonedebounce = 2
Description = Fire
ignore_after = 60
mp3_emails = "JohnD"<xxxxxxxxxx@vzwpix.com>,"Grt Sheriff"<xxxxxxxxxx@vzwpix.com>,"Kevin Chief"<xxxxxxxxxx@vzwpix.com>
Log.txt
uccesfully retrieved local tone info
Succesfully retrieved local tone info
Tone1
Longtone1
defaultdict(<type 'list'>, {'Fire': array([ 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7,
433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7,
433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7,
433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7, 433.7,
433.7, 433.7, 433.7]), 'Amb': array([ 368.5, 368.5, 539. , 539. , 539. , 539. , 539. , 539. ,
539. , 539. , 539. , 539. , 539. , 539. , 539. , 539. ,
539. ])})
defaultdict(<type 'list'>, {'Fire': 1, 'Amb': 0})
247
516
1064
274
494
962
387
360
360
457
838
387
666
252
424
306
768
I can't seem to reproduce this issue. I tried running with your parameters and I'm only getting a single email. Can anyone else reproduce this?
Andy
Found the issue with this. I had hard-coded a 2% tolerance into v59 for testing and forgot to change it back to being configurable. I just posted a v59a that has a fix for this that should give the correct tolerance.
Andy
kenmaples
Member
Thank you! The new version is up and running. I will report back after a few tones come in.
I apologize if this has been previously asked and answered-
Is there a way to configure TTD to "listen" to the audio output feed going to the computer "speakers" rather than the mic or line input?
In other words, can you configure TTD to monitor a streaming feed from the internet? I have thought about using a patch cable to hardwire the speaker output back into the line input, but I wonder if there is a way to "switch" the audio source in software settings.
Thanks
Is there a way to configure TTD to "listen" to the audio output feed going to the computer "speakers" rather than the mic or line input?
In other words, can you configure TTD to monitor a streaming feed from the internet? I have thought about using a patch cable to hardwire the speaker output back into the line input, but I wonder if there is a way to "switch" the audio source in software settings.
Thanks
It follows the record properties of your sound card so it should be able to. What operating system are you using?
jjccoonn
Member
tones editor
Would it be possible to add a print feature into the tone editor part, along with the ability to change entries without having to add and delete to make change's to address.
I'm also wondering if anybody is using follow and for what?
alert_command = c:\test.bat ;Optional parameter. Command to run immediately after the tone is detected
post_email_command = c:\test2.bat ;Optional parameter. Command to run after tone has been detected and emails have been sent
Thanks
J
Would it be possible to add a print feature into the tone editor part, along with the ability to change entries without having to add and delete to make change's to address.
I'm also wondering if anybody is using follow and for what?
alert_command = c:\test.bat ;Optional parameter. Command to run immediately after the tone is detected
post_email_command = c:\test2.bat ;Optional parameter. Command to run after tone has been detected and emails have been sent
Thanks
J
Similar threads
