This is an idea I've had bouncing around for a few years now, and I figured i might as well go ahead and start writing things down and may even start building this. About a year ago, my sister gifted me two old text to speech boxes from the 80s. These are Accent units that work on RS232 in and 12v (regulated to 5v) and will speak any text or numbers sent to them, within reason. They are all hardware based and use COTS-available (for the time) chips for speech synthesis. I think the only programmable chips in the units are eeproms, and even there I may be mistaken. They can speak individual words (american english) or phonemes (for other languages, though I don't know how to do this yet.) The voice is male and heavily distorted. The units were originally designed for disabled individuals who lost the ability to speak and could only communicate through a terminal. I'll probably end up using software TTS on the control PC, but I'd like to at least try and use these boxes for the cool factor. I'll reconsider that when operating this off the grid, as I know for a fact that they are not power efficient, not even remotely.

My idea was to connect one to my baofeng uv-5r 2m/70cm HT for use as a beacon. A standalone PC or RPi (or similar) could send data to the TTS box, then activate PTT on the radio and have the TTS line out run directly to the mic in on the uv-5r. This could transmit my call alone, or date, time, weather information, etc. I'd like it to be able to work entirely off the grid (e.g. no internet or line power), and so far I think this is doable. I have a number of SLA batteries I could dedicate to this, as well as a 15w solar panel. The software TTS options I'm considering don't require network access either, so that shouldn't be a problem.

As I alluded to above, there are a few ways this could be deployed. I'll lay these out here as different options. They wont be too complex but I'd like to keep them separate.

RPi, UV-5r, hardware TTS box, hardwired or solar.

Pros:

• Cool

Cons:

• inefficient
• bulky
• not entirely sure how reliable those hardware TTS boxes are…

RPi w/ software TTS, UV-5r, hardwired or solar.

Pros:

• less power consumption
• physically much smaller

Cons:

• not as cool

• Good info on TTS for Linux: http://elinux.org/RPi_Text_to_Speech_(Speech_Synthesis)
• PiFm: https://github.com/rm-hull/pifm
• Morse Code Wav Generator: https://coderwall.com/p/kotsja/generate-morse-code-wav-files-on-the-linux-cli

• If the hardware TTS is used:
  • Power, size, lifetime of the parts is bad conditions (e.g. heat, humidity)
  • Audio from the hTTS → 5r might be challenging, don't want to blow anything out
• antenna placement/type used (probably the smaller larsen antenna, need a static mount for it)
• need to find a good wiring diagram for PTT & audio

• Antenna!
• Figure out if my solar charge controller can handle two 12v SLAs in series
• Case (probably one of those heavy plastic utility boxes)
• Test out software TTS options. picotts gets good reviews but I don't want it to sound too good, festival is my go-to as long as it's fast enough.
• What am I going to broadcast?
  • call, time, date, temperature, maybe tomorrow's forecast?