chore: add documentation RI

Code functionality left unchanged, only added docs where missing

close: N25B-298

src/robot_interface/endpoints/audio_sender.py

@@ -14,6 +14,24 @@ logger = logging.getLogger(__name__)
 
 
 class AudioSender(SocketBase):
+    """
+    Audio sender endpoint, responsible for sending microphone audio data.
+
+    :param zmq_context: The ZeroMQ context to use.
+    :type zmq_context: zmq.Context
+
+    :param port: The port to use.
+    :type port: int
+
+    :ivar thread: Thread used for sending audio.
+    :type thread: threading.Thread | None
+
+    :ivar audio: PyAudio instance.
+    :type audio: pyaudio.PyAudio | None
+
+    :ivar microphone: Selected microphone information.
+    :type microphone: dict | None
+    """
     def __init__(self, zmq_context, port=settings.agent_settings.audio_sender_port):
         super(AudioSender, self).__init__(str("audio"))  # Convert future's unicode_literal to str
         self.create_socket(zmq_context, zmq.PUB, port)
@@ -30,7 +48,10 @@ class AudioSender(SocketBase):
     def start(self):
         """
         Start sending audio in a different thread.
+
+        Will not start if no microphone is available.
         """
+
         if not self.microphone:
             logger.info("Not listening: no microphone available.")
             return
@@ -41,14 +62,18 @@ class AudioSender(SocketBase):
 
     def wait_until_done(self):
         """
-        Wait until the audio thread is done. Will only be done if `state.exit_event` is set, so
-        make sure to set that before calling this method or it will block.
+        Wait until the audio thread is done.
+
+        Will block until `state.exit_event` is set. If the thread is not running, does nothing.
         """
         if not self.thread: return
         self.thread.join()
         self.thread = None
 
     def _stream(self):
+        """
+        Internal method to continuously read audio from the microphone and send it over the socket.
+        """
         audio_settings = settings.audio_config
         chunk = audio_settings.chunk_size # 320 at 16000 Hz is 20ms, 512 is required for Silero-VAD