Voice alerts test build

Scroll down in music menu. It will first show all switches, then aux knobs if available
and finally the telemetry alarms 1-6. If you do not specify a custom alarm for telemetry alarms, it will play the global telemetry alert mp3 (0013-0018). Values are automatically read out.

Uploaded new builds and voice sets

Nothing substantial happening for you, but there have been quite some internal changes. music.map is history and replaced by the new media/voice.ini as described above. Currently global messages will have ids 0-160 reserved, but actually it won't break anything if you put an id below 161 in the custom section of the ini, it just doesn't make any sense

Please note, as everything has been renamed you will need to update your hardware.ini (voice=dfplayer for most people) and also the corresponding section in your model inis will have to be renamed from "music" to "voice" if you do not want to start over defining things in the voice menu (formerly known as music menu).

edit: I also uploaded a custom devo7e build including voice alert support but without standard mixer GUI to have enough rom space. There is not enough room to include the voice menu, but all entries in model configs will be parsed and should work.

edit2: updated builds again. I thought it might be useful and it was really quick to do, so I also added the option to set custom timer voice messages via the menu

So finally you left only 10 free messages for the global section. It is really a minimum. Begin custom messages at 200 as in one of your preceding posts would have left more room and begin the custom messages on a round figure.

@moeder,

does the new media/voice.ini mean I have to create a new library dfu for my 12s ?
or is it possible to create a /media folder with the voice.ini in it ?

Thanks.

Thanks for the positive feedback! Trying to answer all questions:

1. I will let custom MP3s start at 200 in future builds.
2. those **** welcome and goodbye messages...Since this is no real issue, it still is very far behind on my to do list
3. concerning volume setting: I guess the easiest solution would be to resend the volume command every time the voice queue is finished.
4. concerning channel output as switch: MWM already wrote the "beep" mux for complex mixers, but I never really had a chance to try this out. I didn't change the associated code though, if this works to your liking then we could enhance that. Or do you prefer adding Channel 1-... to the voice menu (which now seems to get pretty full by the way...)
5. more custom voices: Please feel free to add messages to the genspeech standard set I made so far - all files in my github repo are up to date. Either refork and make a pull request or post/pm the desired messages and I will add them. You can always use one of the many online TTS solutions, save the file and add an entry to voice.ini if you wish. I don't know what free tts solutions there are for windows, but I guess we could write a genspeech windows script using espeak as well. But my windows scripting times are long gone...
6. voice and vibration: that would be pretty simple actually...but I guess we should give users the option to select which voice activations they want to have vibration or not.

This all leads back to the problem that the current voice menu design doesn't leave enough room to have all those options in a easy to use manner. I'm open for menu redesign solutions. But I could also think that if this stuff works out well we could make a first push to nightlies and hopefully get some more community action on the code as well.

Nice box you made there, FDR, not much wasted space! Do you velcro it to the TX? But with the low price tag on deployers, there should be one INSIDE every one of your devos

pulpfiction15 wrote: @moeder,

does the new media/voice.ini mean I have to create a new library dfu for my 12s ?
or is it possible to create a /media folder with the voice.ini in it ?

Thanks.

Honestly, as I do not have an 12S and never looked into how it's library handling works, I cannot answer your question.

I fooled around with the mux beep and found it to be fairly simple to use. I preserved its function and added a mux voice with same usage with a channel list added to the voice menu. Once it is bug free I will update builds...probably tomorrow

FDR wrote:
BTW this is my external DFPlayer box:

I like the compact box and the big battery! I am glad to hear that the speakers are performing well for all of us... I was a bit concerned at first, but now for the most part I keep the volume at only about 18 (about 1/2 max).

The box I 3d print is somewhat bigger, and the battery is smaller... I found in the past week of testing, maybe 30-40 minutes flight time with a couple of small quads, the 150mah battery is down from 4.14v to 3.95v -- I'm using a battery from my E010. So I'm thinking battery life of about an hour of regular use with this little battery.


This box is bigger and it allows access to the SD either physically through a side panel or by using the inbuilt micro USB port (my PC sees the SD card as a thumb drive). And I added buttons for volume. The final version includes handles to attach with velcro, but i'm not convinced that is the best way yet... I can't get it to sit right on my 12e. I'm waiting for a 3-pack of dfplayers to put inside the Devo anyway. The box is available on Thingiverse .


Here's the basic circuit board I used. I can provide the fritzing file if anyone wants it, although I have made few modifications to this. It's really a hack job but it works. I think it probably could use a few caps in strategic locations, but since i'm using the Pololu step-up/step-down regulator I'm not worried about power problems. The rest of the parts I had lying around. I would like to have done something with a battery charging system, so it never has to be taken apart, but that's not something i have any experience with... if anyone has suggestions or can help with design, PM me.

Channel voice feature is up and running, I'm just tinkering around with the option of hiding channels in voice menu which do not have a voice mux defined in a mixer. We could do the same for telemetry, so only active telemetry alarms will show up in the menu.

Volume: I can do so. Currently the only other player supported is Audiofx, but this one doesn't offer direct volume setting but only setting volume up and down in 255 steps unfortunately.

Moeder wrote:
Concerning documentation:
The english manual is on github . You need to register at github (free), fork the repository, do changes and make a pull request for merging them. Documentation changes are easy enough to be made on the web interface and the markup language used is pretty self explanatory when looking at existing manual chapters.
There hasn't been a french manual since 2.0, so if you are really serious about doing a good and complete translation, we could extend the github repository to do translated builds as well (FDR, do you hear me ? )

As mentionned I agree to try to make a french translation of the manual but please give me a more detailed instructions in order to do this.

Can someone please clarify exactly what files need to go where on the SD card end in the transmitter and link to them please.

I have downloaded and tried every file attached to posts in this voice alerts forum. My dfplayer is working Standalone just fine, powered by a 5v pololu. At one point the tx stopped beeping, thought I had it, but no. I have literally put each of the files that are unique to the voice alerts in each folder and still nothing.

Also, someone mentioned that the latest nightlys have voice suport, I did not experience that when I just tried the latest.

Running a 12s, and usually get on just fine in these kinds of things. But this one, 9 hours later, I can't do it anymore, clear, concise info would be greatly appreciated.

Thanks in advance
T

On the SD cars all you need to copy to it is the MP3 folder...Be sure to copy the music.map file to the root on the 12s Also you have to enable extended audio in the hardware.ini file on the 12s..

voice >> www.deviationtx.com/downloads-new/catego...o12-v5-0-0-voice-zip

mp3 files>> www.deviationtx.com/downloads-new/catego...oad=4991:voiceen-zip

Ok, after a little bit of code cleanup to reflect the whole change from "music" to "voice" within the code as well I uploaded new builds and voice sets in the usual location (except for standard devo7e build...but no one downloaded it anyways )

Changelog:

• merge in current upstream code
• reduce audio volume setting to 0-10 (it will be internally rescaled to 0-30) to match beep volume setting
• volume command will be resent every time the voice queue is finished, in case it get's lost somewhere along the way
• custom voices now start at ID 200 as requested
• added channel mux "voice" feature: it works exactly as the mux "beep" described by MWM on the first page of this thread (follow the link). Here's a short simple how to get started: set up a complex mixer on a channel, for example with mixer page 1 as mux replace src THR 1-to-1 and set mixer page 2 as mux voice, then select fixed and set scale to the value the channel has to be at to trigger the voice file, for example "0" for a mid stick throttle warning. Don't forget to go to voice menu and define a voice file to be played for a mux voice on the desired channel, or nothing will happen. Try it out, it sounds complex but is really simple and offers plenty of new things to do - thanks FDR for forcing me to realize it sooner than planned
• Code cleanup

Known bugs:

• Welcome and Goodbye messages not always working

The only bigger thing left in my mind is the option to have selectable additional vibration alarm but to be honest I'd prefer to make the big launch into nightlies and maybe have more people work on the code.

@Morlacus: The whole documentation thing should probably be moved into a new thread, as it is not voice related and should get more peoples attention.

@everyone: once merged in to master repo please add "standard" messages to be generated by the genspeech tool so I or anybody else can easily provide nice voice file sets.

FDR wrote: Actually there is no more a music.map file in the root, but a voice.ini in the media dir.
Also the hardware.ini has been changed to have a voice setting instead of the previous extended-audio.

Where is the media dir at i don't have anything named like that on my 12s? Or does it go on the SD card?