Clarification on Running a Python Script in NanoLLM Docker Image

[hasithdd]

Hi, I am a new to jetson andcurrently working on implementing an end-to-end speech model using NanoLLM on an Orin NX 8GB device running JetPack 6.2.
While following the documentation, I couldn’t find a clear example of where to place and execute a Python script. As a test, I ran a test.py script inside the NanoLLM Docker image at /opt/voice_assistant/test.py. Below is the example code. : "import os
import json
from typing import List, Dict, Any, Optional
from datetime import datetime
import numpy as np

from nano_llm import NanoLLM, ChatHistory, BotFunctions, bot_function
from nano_llm.agents.voice_chat import VoiceChat

Define custom bot functions for company-specific tasks

@bot_function
def GET_DATE():
"""Returns the current date."""
return datetime.now().strftime("%A, %B %-d %Y")

@bot_function
def GET_TIME():
"""Returns the current time."""
return datetime.now().strftime("%-I:%M %p")

class CompanyVoiceAssistant:
def init(self,
model_name: str = "meta-llama/Llama-3.2-1B",
quantization: str = None,
api: str = "mlc",
device: str = "cuda"):
"""
Initialize the company voice assistant.

    Args:
        model_name: Name of the LLM model to use
        quantization: Quantization method for the model
        api: Backend API for model inference
        device: Device to run inference on ('cuda' or 'cpu')
    """
    print(f"Initializing voice assistant with {model_name}...")
    
    # Initialize the LLM model
    self.model = NanoLLM.from_pretrained(
        model=model_name,
        quantization=quantization,
        api=api,
        device=device
    )
    
    # Create function calling system
    self.functions = BotFunctions()
    
    # Create system prompt with function documentation
    self.system_prompt = """
    You are a helpful company assistant that can answer questions about the company.
    Be concise and professional in your responses.
    """ + BotFunctions.generate_docs()
    
    # Create voice chat agent with custom STT and TTS
    self.voice_chat = VoiceChat(
        model=self.model,
        stt="whisper_trt",         # Use whisper_trt for speech recognition
        tts="piper",               # Use PiperTTS for speech synthesis
        system_prompt=self.system_prompt,
        functions=self.functions,
        hotword="assistant",       # Wake word to activate the assistant
        continuous=True,           # Keep listening for commands
        vad=True                   # Voice activity detection
    )

def run(self):
    """Start the voice assistant"""
    print("Starting company voice assistant...")
    print("Say 'assistant' to activate, then ask your question")
    self.voice_chat.run()

if name == "main":
import argparse

parser = argparse.ArgumentParser(description="Company Voice Assistant")
parser.add_argument("--model", type=str, default="meta-llama/Llama-3.2-1B", help="LLM model name")
parser.add_argument("--quantization", type=str, default=None, help="Model quantization")

args = parser.parse_args()

assistant = CompanyVoiceAssistant(
    model_name=args.model,
    quantization=args.quantization
)

assistant.run()"