diff --git a/src/auracast/multicast.py b/src/auracast/multicast.py
index ef0303e..29f0eea 100644
--- a/src/auracast/multicast.py
+++ b/src/auracast/multicast.py
@@ -49,9 +49,12 @@ from bumble.audio import io as audio_io
 
 from auracast import auracast_config
 from auracast.utils.read_lc3_file import read_lc3_file
-from auracast.network_audio_receiver import NetworkAudioReceiverUncoded
+from auracast.utils.network_audio_receiver import NetworkAudioReceiverUncoded
+from auracast.utils.webrtc_audio_input import WebRTCAudioInput
 
 
+# Instantiate WebRTC audio input for streaming (can be used per-BIG or globally)
+
 # modified from bumble
 class ModWaveAudioInput(audio_io.ThreadedAudioInput):
     """Audio input that reads PCM samples from a .wav file."""
@@ -357,6 +360,20 @@ class Streamer():
                 big['encoder'] = encoder
                 big['precoded'] = False
 
+            elif audio_source == 'webrtc':
+                big['audio_input'] = WebRTCAudioInput()
+                encoder = lc3.Encoder(
+                    frame_duration_us=global_config.frame_duration_us,
+                    sample_rate_hz=global_config.auracast_sampling_rate_hz,
+                    num_channels=1,
+                    input_sample_rate_hz=48000, # TODO: get samplerate from webrtc
+                )
+                lc3_frame_samples = encoder.get_frame_samples()
+                big['pcm_bit_depth'] = 16
+                big['lc3_frame_samples'] = lc3_frame_samples
+                big['encoder'] = encoder
+                big['precoded'] = False
+
             # precoded lc3 from ram
             elif isinstance(big_config[i].audio_source, bytes):
                 big['precoded'] = True
@@ -470,7 +487,7 @@ class Streamer():
                     if lc3_frame == b'': # Not all streams may stop at the same time
                         stream_finished[i] = True
                         continue
-                else:
+                else: # code lc3 on the fly
                     pcm_frame = await anext(big['audio_input'].frames(big['lc3_frame_samples']), None)
 
                     if pcm_frame is None: # Not all streams may stop at the same time
@@ -483,10 +500,10 @@ class Streamer():
 
                 await big['iso_queue'].write(lc3_frame)
 
-            if all(stream_finished): # Take into account that multiple files have different lengths
-                logging.info('All streams finished, stopping streamer')
-                self.is_streaming = False
-                break
+            # if all(stream_finished): # Take into account that multiple files have different lengths
+            #     logging.info('All streams finished, stopping streamer')
+            #     self.is_streaming = False
+            #     break
 
 
 # -----------------------------------------------------------------------------
diff --git a/src/auracast/server/multicast_frontend.py b/src/auracast/server/multicast_frontend.py
index ed3af61..17ce1cd 100644
--- a/src/auracast/server/multicast_frontend.py
+++ b/src/auracast/server/multicast_frontend.py
@@ -40,9 +40,9 @@ if audio_mode in ["Webapp", "Network"]:
                     name=stream_name,
                     program_info=f"{stream_name} {quality}",
                     language=language,
-                    audio_source="network" if audio_mode=="Webapp" else "network",
+                    audio_source="webrtc" if audio_mode=="Webapp" else "network",
                     input_format="auto",
-                    iso_que_len=64, # TODO: this should be way less to decrease delay
+                    iso_que_len=1, # TODO: this should be way less to decrease delay
                     sampling_frequency=q['rate'],
                     octets_per_frame=q['octets'],
                 ),
diff --git a/src/auracast/server/multicast_server.py b/src/auracast/server/multicast_server.py
index 346a169..36e5a7f 100644
--- a/src/auracast/server/multicast_server.py
+++ b/src/auracast/server/multicast_server.py
@@ -13,6 +13,7 @@ import av
 import av.audio.layout
 from typing import List, Set
 import traceback
+from auracast.utils.webrtc_audio_input import WebRTCAudioInput
 
 app = FastAPI()
 
@@ -31,7 +32,6 @@ global_config_group = auracast_config.AuracastConfigGroup()
 # Create multicast controller
 multicaster: multicast_control.Multicaster | None = None
 
-
 @app.post("/init")
 async def initialize(conf: auracast_config.AuracastConfigGroup):
     """Initializes the broadcasters."""
@@ -126,6 +126,7 @@ async def offer(offer: Offer):
     # create directory for records - only for testing
     os.makedirs("./records", exist_ok=True)
 
+    multicaster.start_streaming()
     @pc.on("track")
     async def on_track(track: MediaStreamTrack):
         log.info(f"{id_}: track {track.kind} received")
@@ -145,13 +146,25 @@ async def offer(offer: Offer):
                     log.info(f"array.shape{frame_array.shape}")
                     log.info(f"array.dtype{frame_array.dtype}")
                     log.info(f"frame.to_ndarray(){frame_array}")
+
+                    samples = frame_array.reshape(-1)
+                    log.info(f"samples.shape: {samples.shape}")
+
+                    if frame.layout.name == 'stereo':
+                        # Interleaved stereo: [L0, R0, L1, R1, ...]
+                        mono_array = samples[::2]  # Take left channel
+                    else:
+                        mono_array = samples
+
+                    log.info(f"mono_array.shape: {mono_array.shape}")
+
+                    audio_input = list(multicaster.bigs.values())[0]['audio_input']
+
                     
                 frame_array = frame.to_ndarray()
-                log.info(f"frame_array.shape: {frame_array.shape}")
 
                 # Flatten in case it's (1, N) or (N,)
                 samples = frame_array.reshape(-1)
-                log.info(f"samples.shape: {samples.shape}")
 
                 if frame.layout.name == 'stereo':
                     # Interleaved stereo: [L0, R0, L1, R1, ...]
@@ -159,7 +172,8 @@ async def offer(offer: Offer):
                 else:
                     mono_array = samples
 
-                log.info(f"mono_array.shape: {mono_array.shape}")
+                # Feed mono PCM samples to the global WebRTC audio input
+                await audio_input.put_samples(mono_array.astype(np.int16))
 
                 # Save to WAV file - only for testing
                 # if not hasattr(pc, 'wav_writer'):
@@ -172,9 +186,8 @@ async def offer(offer: Offer):
 
                 # pcm_data = mono_array.astype(np.int16).tobytes()
                 # pc.wav_writer.writeframes(pcm_data)
-                
 
-                
+
         except Exception as e:
             log.error(f"{id_}: Exception in on_track: {e}")
         finally:
@@ -218,7 +231,7 @@ async def shutdown():
 if __name__ == '__main__':
     import uvicorn
     log.basicConfig(
-        level=log.INFO,
+        level=log.DEBUG,
         format='%(module)s.py:%(lineno)d %(levelname)s: %(message)s'
     )
     uvicorn.run(app, host="0.0.0.0", port=5000)
\ No newline at end of file
diff --git a/src/auracast/network_audio_receiver.py b/src/auracast/utils/network_audio_receiver.py
similarity index 100%
rename from src/auracast/network_audio_receiver.py
rename to src/auracast/utils/network_audio_receiver.py
diff --git a/src/auracast/utils/webrtc_audio_input.py b/src/auracast/utils/webrtc_audio_input.py
new file mode 100644
index 0000000..0e6b1a4
--- /dev/null
+++ b/src/auracast/utils/webrtc_audio_input.py
@@ -0,0 +1,41 @@
+import asyncio
+import numpy as np
+import logging
+
+class WebRTCAudioInput:
+    """
+    Buffer PCM samples from WebRTC and provide an async generator interface for chunked frames.
+    """
+    def __init__(self):
+        self.buffer = np.array([], dtype=np.int16)
+        self.lock = asyncio.Lock()
+        self.data_available = asyncio.Event()
+        self.closed = False
+
+    async def frames(self, frame_size: int):
+        """
+        Async generator yielding exactly frame_size samples as numpy arrays.
+        """
+        while not self.closed:
+            async with self.lock:
+                if len(self.buffer) >= frame_size:
+                    chunk = self.buffer[:frame_size]
+                    self.buffer = self.buffer[frame_size:]
+                    logging.debug(f"WebRTCAudioInput: Yielding {frame_size} samples, buffer now has {len(self.buffer)} samples remaining.")
+                    yield chunk
+                else:
+                    self.data_available.clear()
+            await self.data_available.wait()
+
+    async def put_samples(self, samples: np.ndarray):
+        """
+        Add new PCM samples (1D np.int16 array, mono) to the buffer.
+        """
+        async with self.lock:
+            self.buffer = np.concatenate([self.buffer, samples])
+            logging.debug(f"WebRTCAudioInput: Added {len(samples)} samples, buffer now has {len(self.buffer)} samples.")
+            self.data_available.set()
+
+    def close(self):
+        self.closed = True
+        self.data_available.set()