Build AI Voice Agent for Banking KYC

1python -m venv venv
2source venv/bin/activate  # On Windows use `venv\Scripts\activate`
3

1pip install videosdk-agents videosdk-plugins-silero videosdk-plugins-turn-detector videosdk-plugins-deepgram videosdk-plugins-openai videosdk-plugins-elevenlabs
2

1VIDEOSDK_API_KEY=your_videosdk_api_key
2

1import asyncio, os
2from videosdk.agents import Agent, AgentSession, CascadingPipeline, JobContext, RoomOptions, WorkerJob, ConversationFlow
3from videosdk.plugins.silero import SileroVAD
4from videosdk.plugins.turn_detector import TurnDetector, pre_download_model
5from videosdk.plugins.deepgram import DeepgramSTT
6from videosdk.plugins.openai import OpenAILLM
7from videosdk.plugins.elevenlabs import ElevenLabsTTS
8from typing import AsyncIterator
9
10pre_download_model()
11
12agent_instructions = "You are a knowledgeable and efficient AI Voice Agent specialized in assisting users with banking KYC (Know Your Customer) processes. Your primary role is to guide users through the KYC process, answer questions related to required documents, and provide information on how to securely submit their information. You can also provide general information about the importance of KYC in banking and how it helps in fraud prevention.\n\nCapabilities:\n1. Explain the steps involved in the KYC process for banking.\n2. List and describe the documents required for KYC verification.\n3. Provide instructions on how to securely upload or submit KYC documents.\n4. Answer frequently asked questions about banking KYC.\n5. Offer insights into the benefits of completing KYC for both the bank and the customer.\n\nConstraints and Limitations:\n1. You are not authorized to collect or store any personal data from users.\n2. You cannot provide legal advice or interpret banking regulations.\n3. Always remind users to verify information with their specific bank as procedures may vary.\n4. Include a disclaimer that users should contact their bank for any specific issues or concerns related to their KYC process."
13
14class MyVoiceAgent(Agent):
15    def __init__(self):
16        super().__init__(instructions=agent_instructions)
17    async def on_enter(self): await self.session.say("Hello! How can I help?")
18    async def on_exit(self): await self.session.say("Goodbye!")
19
20async def start_session(context: JobContext):
21    agent = MyVoiceAgent()
22    conversation_flow = ConversationFlow(agent)
23
24    pipeline = CascadingPipeline(
25        stt=DeepgramSTT(model="nova-2", language="en"),
26        llm=OpenAILLM(model="gpt-4o"),
27        tts=ElevenLabsTTS(model="eleven_flash_v2_5"),
28        vad=SileroVAD(threshold=0.35),
29        turn_detector=TurnDetector(threshold=0.8)
30    )
31
32    session = AgentSession(
33        agent=agent,
34        pipeline=pipeline,
35        conversation_flow=conversation_flow
36    )
37
38    try:
39        await context.connect()
40        await session.start()
41        await asyncio.Event().wait()
42    finally:
43        await session.close()
44        await context.shutdown()
45
46def make_context() -> JobContext:
47    room_options = RoomOptions(
48    #  room_id="YOUR_MEETING_ID",  # Set to join a pre-created room; omit to auto-create
49        name="VideoSDK Cascaded Agent",
50        playground=True
51    )
52
53    return JobContext(room_options=room_options)
54
55if __name__ == "__main__":
56    job = WorkerJob(entrypoint=start_session, jobctx=make_context)
57    job.start()
58

1pipeline = CascadingPipeline(
2    stt=DeepgramSTT(model="nova-2", language="en"),
3    llm=OpenAILLM(model="gpt-4o"),
4    tts=ElevenLabsTTS(model="eleven_flash_v2_5"),
5    vad=SileroVAD(threshold=0.35),
6    turn_detector=TurnDetector(threshold=0.8)
7)
8

1session = AgentSession(
2    agent=agent,
3    pipeline=pipeline,
4    conversation_flow=conversation_flow
5)
6
7try:
8    await context.connect()
9    await session.start()
10    await asyncio.Event().wait()
11finally:
12    await session.close()
13    await context.shutdown()
14

1python main.py
2