I did post some links earlier, where PPT had been done with LoRa module SX1278 on 446MHz @ 1600bps codec2. Furthermore, as discussed before, we could use different channel parameters for PPT and mesh data, to gain better audio capabilities (with shorter audio range) and longer mesh data capabilities (with slower transfer speeds).
Although, I like the idea of having both PPT audio flood (446MHz) and mesh data transfer capabilities (433MHz) on one device, I think we shall concentrate on solid mesh data capabilities first.
The trick to avoid duty cycle limits is to operate on PMR 446MHz for PPT (not on LoRa 433MHz).
I think we’re saying the same thing. PPT over LoRa won’t be usable. Switch to something else.
Perhaps I was being to gentle in suggesting there are latency and bandwidth issues.
It has been done in the past with audio recordings and medium-range setup, I don’t see why it shouldn’t be a possible feature to develop in the future.
What if we try to support short audio messages? Like most chat apps do.
Could be nice indeed. However, the problem concerning the low data bandwidth capabilities of LoRa, combined with high data bandwidth needs of voice transmissions, has to be resolved first. As @sens8tion said; it all comes to the question of how long would people be happy waiting for 10 seconds of audio.
There are some efforts of other groups regarding this, although it’s not an easy process! I wouldn’t hold my breath, for adding a voice feature, at least at this stage.
If no one cared about robot speak. A voice to text, transmit text, text to speech may work. My android speaks my text and converts my speech to text.
Besides the long communication delay, are there any technical challenges in sending files over LoRa?
258 byte limit so you would have to have a way to breakup and reconstruct the files.
I don’t understand why some people here try to turn Lora mesh into PTT voice radio!
You don’t like $30 (Baofeng) radios or more expensive (Motorola) radios any more?
Then the idea of sending text on 433Mhz, and transmitting voice to 446Mhz is useless.
You want to transmit voice at 446Mhz use a $30 Baofeng and to send messages with the Meshtastic use Lora radios with the mesh network.
To broadcast at 446Mhz also a license is needed, which is useless for people that does not have a license.
Here it seems that HAM’s don’t like their licensed radios anymore, they want to invade even a system that doesn’t need one.
@TitanTronics You are not wrong…
As to your question, I think people are looking to leverage the extended range and reliability of the mesh technology.
That said, personally, when disagreeing with people I try to do so as politely as I can mange.
People like Mesh technology and it’s a great thing. But when I see HAMs pretending they don’t understand how radio transmissions works and their transmission speed, it amazes me how much misinformation there is among HAMs.
Lora devices use 20mW ham radios use 1kW of power.
I believe their 1kW antennas and transmissions can communicate over 2000km in full reliability.
What reliability are they looking for with the 20mW of lora devices.
Instead of seeing network mesh and Meshtastic as one more tool in their arsenal they just want to convert it to HAM, they amaze me more and more.
You are completely wrong about HAM community and their radios. The radio amateurs use and QRP ( low power transmissions) both for voice and data(analog & digital) to any band without any repeater devices, but because technology goes forward their interested for lora. Please do not criticize them.
Yes, of course I am wrong because 2W ~ 20w / 0.5~30Mhz HF for HAM’s are not enough, do you know how much distance you can cover with that power?
Calling 20W QRP radios, low power is ridiculous. With 20Watt HF QRP radios you can cover huge distances without any need for mesh or repeaters
And don’t complain because you have the power to cover all distances without the need for Lora.
If you want the mesh use Lora with Meshtastic, you will have fun even without the QRP Radios
There have been successful attempts to transmit audio over LoRa using reasonable transmit power ~1W and high gain directional Yagi antennae.
Have a video or post to share about such attempts @Anelito?
VHF meshes are expensive and are subject to strict regulations in most countries whereas ISM bands are quite commonly free to use. Moreover, HAM radios cannot use encryption.
Zello PTT app is using SpeedX. Speex is an Open Source/Free Software patent-free audio compression format designed for speech. As far as I know, max 15 sec messages and we are taking about byte size packages.
I think the main issue with attempting a voice service right now is the impact on duty cycle and the limitations of the current flood mesh implementation.
Once the mesh layer had been upgraded PPT Voice, perhaps limited to the shorter range / faster channels, would be much more likely without completely disregarding duty cycle limits.
That said, people are starting to create plugins, and I’m sure this would be a popular one. Just know, 4kbps is still a lot of bandwidth for LORA
This could be an even better codec option, requiring 1/4 the bandwidth. Codec 2 700C – Rowetel
I have some friends at APC who are in touch with Rowe, the Codec2 creator, and have offered to connect him to anyone interested in implementing it with Meshtastic. Let me know if someone’s interested.
A plugin for voice would be amazing!
Plugin for voice would be easy but there would be hardware requirements to make that work. The hardware bits also won’t be too difficult, but that adds a barrier to entry.
Maybe I’ll look into it after I’m done with my next plugin. The one I’m working on now isn’t easy, so it’ll be a while.
How ever way it works out, codec 2 has already been ported to the esp32 using the Arduino framework, so this won’t be too difficult. So we can take that as a proof of concept.
That site provides this diagram for the proof of concept:
If this where to be implemented with Meshtastic in mind, probably will want to add the microphone (so there won’t be a need for a separate board), speaker amplifier circuit (super easy), a button, a few status LEDs and maybe a separate power regulator.
If anyone would want to work on it, I’d gladly be a sounding board.