return numpy output of audio

app.py

@@ -7,7 +7,6 @@ from vocode import getenv
 import gradio as gr
 import os
 import logging
-import sounddevice as sd
 from vocode.turn_based.agent.chat_gpt_agent import ChatGPTAgent
 from vocode.turn_based.synthesizer import CoquiSynthesizer
 from vocode.turn_based.transcriber.whisper_transcriber import WhisperTranscriber
@@ -88,16 +87,12 @@ def convert_to_audio_segment(input_audio):
 def send_audio(audio_segment: AudioSegment):
     logger.info("now processing output")
     sampling_rate = DEFAULT_SAMPLING_RATE
-    stream = sd.OutputStream(
-        channels=1,
-        samplerate=sampling_rate,
-        dtype=np.int16,
-        device=None,
-    )
     raw_data = audio_segment.raw_data
     if audio_segment.frame_rate != sampling_rate:
         raw_data = audio_segment.set_frame_rate(sampling_rate).raw_data
-    return stream.write(np.frombuffer(raw_data, dtype=np.int16))
+    output = np.frombuffer(raw_data, dtype=np.int16)
+
+    return output
 
 
 def main(input_audio):