Build a Speech-to-Text AI Agent in Python

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

1pip install videosdk
2pip install python-dotenv
3

1VIDEOSDK_API_KEY=your_api_key_here
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
10# Pre-downloading the Turn Detector model
11pre_download_model()
12
13agent_instructions = "You are a 'Speech to Text AI Agent' developed using Python, integrated within the VideoSDK framework. Your primary role is to assist users by converting spoken language into written text accurately and efficiently. You are designed to be a helpful and efficient transcription assistant.\n\nCapabilities:\n1. Convert spoken language into text in real-time with high accuracy.\n2. Support multiple languages and dialects, adapting to various accents.\n3. Provide users with options to edit and save transcriptions.\n4. Handle background noise and differentiate between multiple speakers when possible.\n5. Offer transcription summaries and keyword highlights upon request.\n\nConstraints and Limitations:\n1. You are not capable of understanding or interpreting the content beyond transcription.\n2. You must inform users that the transcription accuracy may vary based on audio quality and clarity.\n3. You cannot store or share any transcriptions without explicit user consent.\n4. You must include a disclaimer that the transcriptions are for informational purposes only and should be verified for accuracy.\n5. You are limited to processing audio inputs of up to 60 minutes in length per session."
14
15class MyVoiceAgent(Agent):
16    def __init__(self):
17        super().__init__(instructions=agent_instructions)
18    async def on_enter(self): await self.session.say("Hello! How can I help?")
19    async def on_exit(self): await self.session.say("Goodbye!")
20
21async def start_session(context: JobContext):
22    # Create agent and conversation flow
23    agent = MyVoiceAgent()
24    conversation_flow = ConversationFlow(agent)
25
26    # Create pipeline
27    pipeline = CascadingPipeline(
28        stt=DeepgramSTT(model="nova-2", language="en"),
29        llm=OpenAILLM(model="gpt-4o"),
30        tts=ElevenLabsTTS(model="eleven_flash_v2_5"),
31        vad=SileroVAD(threshold=0.35),
32        turn_detector=TurnDetector(threshold=0.8)
33    )
34
35    session = AgentSession(
36        agent=agent,
37        pipeline=pipeline,
38        conversation_flow=conversation_flow
39    )
40
41    try:
42        await context.connect()
43        await session.start()
44        # Keep the session running until manually terminated
45        await asyncio.Event().wait()
46    finally:
47        # Clean up resources when done
48        await session.close()
49        await context.shutdown()
50
51def make_context() -> JobContext:
52    room_options = RoomOptions(
53    #  room_id="YOUR_MEETING_ID",  # Set to join a pre-created room; omit to auto-create
54        name="VideoSDK Cascaded Agent",
55        playground=True
56    )
57
58    return JobContext(room_options=room_options)
59
60if __name__ == "__main__":
61    job = WorkerJob(entrypoint=start_session, jobctx=make_context)
62    job.start()
63

1curl -X POST "https://api.videosdk.live/v1/rooms" \
2-H "Authorization: Bearer YOUR_API_KEY" \
3-H "Content-Type: application/json" \
4-d '{}'
5

1class MyVoiceAgent(Agent):
2    def __init__(self):
3        super().__init__(instructions=agent_instructions)
4    async def on_enter(self): await self.session.say("Hello! How can I help?")
5    async def on_exit(self): await self.session.say("Goodbye!")
6

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

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

1def make_context() -> JobContext:
2    room_options = RoomOptions(
3        name="VideoSDK Cascaded Agent",
4        playground=True
5    )
6    return JobContext(room_options=room_options)
7

1if __name__ == "__main__":
2    job = WorkerJob(entrypoint=start_session, jobctx=make_context)
3    job.start()
4

1python main.py
2