ESPAsyncWebServer replacing AudioWAVServer

[honza540]

Hi Phil,
How can I integrate your library used in my "audio over net" (I2S mic stream accessible over browser) project with ESPAsyncWebServer instead of AudioWAVServer? I need to add to the web page a boost volume control-slider (the boost gain as you helped me solve in the other thread) and authentication, but for that I need a html code to be running and the AudioWAVServer just returns data of the format audio/wav - how can I "redirect" it into the ESPAsyncWebServer entity?
Running the webservers on different ports is not a real option.

Thank you for your ideas or even better code sample :-)

Best regards, Jan

[pschatzmann]

I don't know the ESPAsyncWebServer: you are at the wrong place here!

[honza540]

OK, I see. Could be added the basic authentication user/password to the used AudioWAVServer then?

[LukedeMunk]

I also use ESPAsyncWebServer (for commands) in combination with the AudioWAVServer. I use port 80 for ESPAsyncWebServer and port 81 for AudioWAVServer. Why would seperate ports not work for your project?

[joba-1]

compared to the standard esp webserver implementation the async webserver can do way more requests per second with less latency and without blocking other important tasks. Sounds like s good match for time critical audio tasks. So I would understand the question as „does AudioWavServer already use async algo or can it be rewritten to do this“
Never tried to combi e this myself so I dont know…

[honza540]

the idea is to "secure" it behind the login/password authentication, I do not have any specific webserver preference (ESPAsyncWebServer was just a wild guess what can be used), so in general, I would use anything, what will allow me implement at least very basic security (SSL is already too complex for ESP 32) and also add "listener" for POST or GET (data in URL) requests submitting values necessary for the audio processing (currently only for volume: VolumeStream volumeStream(i2sStream); volumeStream.setVolume(volumeFromWEB);) so that I can configure the audio output remotely.

[LukedeMunk]

Then you could just setup the ESPAsyncWebServer as "command server" on port 80. Make endpoints to configure the audio and login. When a user is logged in (through the "command server"), make the AudioWAVServer ("audio server") available. Then maybe redirect to the "audio server" listening page. Or make a JavaScript Audio object to listen to the "audio server" on the same webpage.

[honza540]

I believe that in the case the authenticated client connected to port 80, then port 81 will be available also for "everyone", also unauthenticated clients, which is probably a dead-end route. I would prefer to integrate both into one web server - authentication and audio streaming.

[honza540]

The choice could the library AudioWAVServer, I just do not know "how to" :-(

[LukedeMunk]

Unfortunately, I cannot help with mixing up the servers. For seperate servers: You could work with IP based sessions, then you whitelist those IPs so that only the authorized clients can listen to the audio on port 81. In terms of security, I believe it is feasible to do it this way.