feat: implement transcriber agent

Uses speech fragments of the VAD agent, emits transcribed text over SPADE's default communication channel to no recipient for now.

ref: N25B-209

src/control_backend/agents/vad_agent.py

@@ -7,6 +7,7 @@ import zmq.asyncio as azmq
 from spade.agent import Agent
 from spade.behaviour import CyclicBehaviour
 
+from control_backend.agents.transcription import TranscriptionAgent
 from control_backend.core.config import settings
 from control_backend.core.zmq_context import context as zmq_context
 
@@ -145,10 +146,13 @@ class VADAgent(Agent):
         if audio_out_port is None:
             await self.stop()
             return
+        audio_out_address = f"tcp://localhost:{audio_out_port}"
 
         streaming = Streaming(self.audio_in_socket, self.audio_out_socket)
         self.add_behaviour(streaming)
 
-        # ... start agents dependent on the output audio fragments here
+        # Start agents dependent on the output audio fragments here
+        transcriber = TranscriptionAgent(audio_out_address)
+        await transcriber.start()
 
         logger.info("Finished setting up %s", self.jid)