[Bug]: turn based conversation playground fails execution

[matteo-ticli]

Brief Description

I am new to the repo and I am trying it out. I have encountered a bug using the following code:

import asyncio
import signal

from pydantic_settings import BaseSettings, SettingsConfigDict

from vocode.helpers import create_streaming_microphone_input_and_speaker_output, \
    create_turn_based_microphone_input_and_speaker_output
from vocode.logging import configure_pretty_logging
from vocode.turn_based.agent.chat_gpt_agent import ChatGPTAgent
from vocode.turn_based.synthesizer.eleven_labs_synthesizer import ElevenLabsSynthesizer
from vocode.turn_based.transcriber.whisper_transcriber import WhisperTranscriber
from vocode.turn_based.turn_based_conversation import TurnBasedConversation

configure_pretty_logging()


class Settings(BaseSettings):
    """
    Settings for the turn-based conversation quickstart.
    These parameters can be configured with environment variables.
    """
    model_config = SettingsConfigDict(
        env_file=".env",
        env_file_encoding="utf-8",
    )

    # Define environment variables
    deepgram_api_key: str
    openai_api_key: str
    elevenlabs_api_key: str


settings = Settings()

if __name__ == "__main__":
    (
        microphone_input,
        speaker_output,
    ) = create_turn_based_microphone_input_and_speaker_output(
        use_default_devices=True,
    )

    conversation = TurnBasedConversation(
        input_device=microphone_input,
        output_device=speaker_output,
        transcriber=WhisperTranscriber(api_key=settings.openai_api_key),
        agent=ChatGPTAgent(
            api_key=settings.openai_api_key,
            system_prompt="The AI is having a pleasant conversation about life",
            initial_message="Hello!",
        ),
        synthesizer=ElevenLabsSynthesizer(
            api_key=settings.elevenlabs_api_key,
            voice_id="Lhvb6ub1gbURNYItrKHe",
            stability=0.75,
            similarity_boost=0.5
        )
    )
    print("Starting conversation. Press Ctrl+C to exit.")
    while True:
        try:
            input("Press enter to start recording...")
            conversation.start_speech()
            input("Press enter to end recording...")
            conversation.end_speech_and_respond()
        except KeyboardInterrupt:
            break

The error message is the following:

Traceback (most recent call last):
  File "/Users/matteoticli/coding_projects/AIphone/AIphone/test_turn_conversation.py", line 47, in <module>
    conversation = TurnBasedConversation(
                   ^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/matteoticli/coding_projects/AIphone/venv/lib/python3.11/site-packages/vocode/turn_based/turn_based_conversation.py", line 24, in __init__
    self.maybe_play_initial_message()
  File "/Users/matteoticli/coding_projects/AIphone/venv/lib/python3.11/site-packages/vocode/turn_based/turn_based_conversation.py", line 28, in maybe_play_initial_message
    self.output_device.send_audio(self.synthesizer.synthesize(self.agent.initial_message))
                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/matteoticli/coding_projects/AIphone/venv/lib/python3.11/site-packages/vocode/turn_based/synthesizer/eleven_labs_synthesizer.py", line 50, in synthesize
    return AudioSegment.from_mp3(io.BytesIO(audio))  # type: ignore
                                 ^^^^^^^^^^^^^^^^^
TypeError: a bytes-like object is required, not 'generator'

How can I resolve this?

LLM

None

Transcription Services

None

Synthesis Services

None

Telephony Services

None

Conversation Type and Platform

No response

Steps to Reproduce

.

Expected Behavior

.

Screenshots

No response