Tag Archives: Speech to Text

AI on the Edge LESSON 36: Select Active Camera in OpenCV With Voice Commands

July 24, 2026 admin

Welcome back, makers, engineers, and AI enthusiasts! In our previous lessons, we built out robust multi-camera setups, streaming feeds from both Raspberry Pi cameras and high-definition USB cameras. But watching four feeds in separate tiles is only half the battle. What happens when you want to interact with your system hands-free?

In this lesson, we are taking our edge vision projects to the next level by integrating voice commands. You will learn how to run a dedicated speech-to-text thread in the background, catch voice triggers safely using a thread-safe queue, and dynamically switch your active main camera feed in OpenCV on the fly—just by speaking.

What You Will Learn in This Lesson

Multi-Threaded Speech Recognition: How to run the stt listener in a separate daemon thread so it never blocks or stutters your high-FPS video processing loop.
Thread-Safe Communication: Using Python’s Queue module to pass voice commands seamlessly from the background listening thread into your main application loop.
Handling STT Variations: Accounting for common speech-to-text homophone variations (like “two”, “to”, and “too”, or “four” and “for”) to make your voice control robust and reliable.
Dynamic Frame Routing: Mapping spoken commands to specific camera streams (piCam1, piCam2, usbCam1, and usbCam2) and updating your primary display window instantly.
Organized Window Layouts: Positioning and sizing multiple OpenCV windows on your desktop workspace for a clean, professional multi-camera dashboard.

Step-by-Step Breakdown of the Script

1. Setting Up the Background Voice Thread

When working with real-time video processing in OpenCV, blocking functions are your worst enemy. If you call a speech recognition listening function directly inside your main while loop, your video frames will freeze while waiting for audio input.

To prevent this, we initialize a background worker function (getCamera) and launch it as a daemon thread:

The thread continuously listens for your voice commands using the Fusion Hat STT module.
Once a command is captured, it strips whitespace, checks for exit triggers (like saying 'quit'), and pushes valid commands directly into our commandQ.

2. Initializing Multiple Cameras

Our setup harnesses the full power of the hardware by combining native Raspberry Pi camera interfaces with standard USB webcams:

Pi Cameras: Configured via Picamera2 at a crisp 1280x720 resolution with RGB888 formatting running smoothly at 60 frames per second.
USB Cameras: Initialized through OpenCV’s VideoCapture class, set to target resolution and optimized for 30 FPS.

3. Processing and Routing Commands in the Main Loop

Inside the primary application loop, our script handles three core tasks simultaneously:

Calculate Performance: Continuously tracks and smooths out the frames-per-second (FPS) metric so you can monitor system load.
Check the Command Queue: Non-blockingly checks if the background thread has dropped a new camera selection into commandQ. If a new command is waiting, it updates the mainCam variable.
Route the Main Frame: Evaluates the active camera string—including clever fallback checks for common voice misinterpretations like 'camera to' or 'camera for'—and assigns the corresponding video stream to mainFrame.

4. Managing Your OpenCV Windows

To give you a complete command-center experience, the script generates a large primary display window for your active camera view, accompanied by a clean row of smaller preview tiles across the bottom of your screen for all four connected feeds.

# ====================================================================
# DISCLAIMER:
# This code is provided as-is for educational and experimental 
# purposes only. The author makes no representations or warranties of 
# any kind concerning the safety, suitability, or accuracy of this 
# code. Use at your own risk. The author assumes no liability for any 
# damages, system failures, security breaches, or network issues 
# resulting from the use or implementation of this script.
# ====================================================================
import cv2
import time
from picamera2 import Picamera2
import threading
from queue import Queue
from fusion_hat.stt import STT
stt = STT('en-us')
commandQ = Queue()
mainCam = 'camera one'

W=1280
H=720
tStart = time.time()
fps = 0
RES = (W,H)

piCam1 = Picamera2(0)
piCam1.preview_configuration.main.size = RES
piCam1.preview_configuration.main.format = "RGB888"
piCam1.preview_configuration.controls.FrameRate=60
piCam1.preview_configuration.align()
piCam1.configure("preview")
piCam1.start()

piCam2 = Picamera2(1)
piCam2.preview_configuration.main.size = RES
piCam2.preview_configuration.main.format = "RGB888"
piCam2.preview_configuration.controls.FrameRate=60
piCam2.preview_configuration.align()
piCam2.configure("preview")
piCam2.start()

usbCam1 = cv2.VideoCapture(16)
usbCam1.set(cv2.CAP_PROP_FRAME_WIDTH,W)
usbCam1.set(cv2.CAP_PROP_FRAME_HEIGHT,H)
usbCam1.set(cv2.CAP_PROP_FPS, 30)

usbCam2 = cv2.VideoCapture(18)
usbCam2.set(cv2.CAP_PROP_FRAME_WIDTH,W)
usbCam2.set(cv2.CAP_PROP_FRAME_HEIGHT,H)
usbCam2.set(cv2.CAP_PROP_FPS, 30)

textLowerLeft = (int(W*.01),int(H*.05))
fontFace = cv2.FONT_HERSHEY_SIMPLEX
fontThickness = int(W/425)
fontScale = H*.0015
fontColor = (0,0,255)

running = True
def getCamera():
    print('Voice Thread Started')
    global running
    while running:
        myCamera = stt.listen(stream = False)
        myCamera = myCamera.strip()
        print(myCamera)
        if myCamera == 'quit':
            print('Security Camera Offline')
            running = False
            break
        commandQ.put(myCamera)
    print('Thread Terminated')

cameraThread = threading.Thread(target = getCamera,daemon=True)
cameraThread.start()

cv2.namedWindow('Main Camera',cv2.WINDOW_GUI_NORMAL)
cv2.moveWindow('Main Camera',0,65)
cv2.resizeWindow('Main Camera',W,H)

cv2.namedWindow('Camera1',cv2.WINDOW_GUI_NORMAL)
cv2.moveWindow('Camera1',0,H)
cv2.resizeWindow('Camera1',int(W/4),int(H/4))

cv2.namedWindow('Camera2',cv2.WINDOW_GUI_NORMAL)
cv2.moveWindow('Camera2',int(W/4),H)
cv2.resizeWindow('Camera2',int(W/4),int(H/4))

cv2.namedWindow('Camera3',cv2.WINDOW_GUI_NORMAL)
cv2.moveWindow('Camera3',2*int(W/4),H)
cv2.resizeWindow('Camera3',int(W/4),int(H/4))

cv2.namedWindow('Camera4',cv2.WINDOW_GUI_NORMAL)
cv2.moveWindow('Camera4',3*int(W/4),H)
cv2.resizeWindow('Camera4',int(W/4),int(H/4))

while running:
    deltaT = time.time() - tStart
    tStart=time.time()
    fps = fps*.95 + (1/deltaT)*.05
    frame1= piCam1.capture_array()
    #frame1=cv2.flip(frame1,-1)
    
    frame2= piCam2.capture_array()
    frame2=cv2.flip(frame2,-1)
    
    _ , frame3 = usbCam1.read()
    _ , frame4 = usbCam2.read()
    mainFrame = frame1
    
    if commandQ.empty() == False:
        mainCam = commandQ.get()
        
    if mainCam == 'camera one':
        mainFrame = frame1
    if mainCam == 'camera two' or mainCam == 'camera to' or mainCam == 'camera too':
        mainFrame = frame2
    if mainCam == 'camera three':
        mainFrame = frame3
    if mainCam == 'camera four' or mainCam == 'camera for':
        mainFrame = frame4
    myText = "FPS: "+str(round(fps,1))
    cv2.putText(frame1,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)
    cv2.imshow("Main Camera", mainFrame)
    cv2.imshow("Camera1", frame1)
    cv2.imshow("Camera2", frame2)
    cv2.imshow("Camera3", frame3)
    cv2.imshow("Camera4", frame4)    
    if cv2.waitKey(1)==ord('q'):
        break
cv2.destroyAllWindows()
print('Program Terminated')

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

# ====================================================================

# DISCLAIMER:

# This code is provided as-is for educational and experimental

# purposes only. The author makes no representations or warranties of

# any kind concerning the safety, suitability, or accuracy of this

# code. Use at your own risk. The author assumes no liability for any

# damages, system failures, security breaches, or network issues

# resulting from the use or implementation of this script.

# ====================================================================

import cv2

import time

from picamera2 import Picamera2

import threading

from queue import Queue

from fusion_hat.stt import STT

stt = STT('en-us')

commandQ = Queue()

mainCam = 'camera one'

W=1280

H=720

tStart = time.time()

fps = 0

RES = (W,H)

piCam1 = Picamera2(0)

piCam1.preview_configuration.main.size = RES

piCam1.preview_configuration.main.format = "RGB888"

piCam1.preview_configuration.controls.FrameRate=60

piCam1.preview_configuration.align()

piCam1.configure("preview")

piCam1.start()

piCam2 = Picamera2(1)

piCam2.preview_configuration.main.size = RES

piCam2.preview_configuration.main.format = "RGB888"

piCam2.preview_configuration.controls.FrameRate=60

piCam2.preview_configuration.align()

piCam2.configure("preview")

piCam2.start()

usbCam1 = cv2.VideoCapture(16)

usbCam1.set(cv2.CAP_PROP_FRAME_WIDTH,W)

usbCam1.set(cv2.CAP_PROP_FRAME_HEIGHT,H)

usbCam1.set(cv2.CAP_PROP_FPS, 30)

usbCam2 = cv2.VideoCapture(18)

usbCam2.set(cv2.CAP_PROP_FRAME_WIDTH,W)

usbCam2.set(cv2.CAP_PROP_FRAME_HEIGHT,H)

usbCam2.set(cv2.CAP_PROP_FPS, 30)

textLowerLeft = (int(W*.01),int(H*.05))

fontFace = cv2.FONT_HERSHEY_SIMPLEX

fontThickness = int(W/425)

fontScale = H*.0015

fontColor = (0,0,255)

running = True

def getCamera():

print('Voice Thread Started')

global running

while running:

myCamera = stt.listen(stream = False)

myCamera = myCamera.strip()

print(myCamera)

if myCamera == 'quit':

print('Security Camera Offline')

running = False

break

commandQ.put(myCamera)

print('Thread Terminated')

cameraThread = threading.Thread(target = getCamera,daemon=True)

cameraThread.start()

cv2.namedWindow('Main Camera',cv2.WINDOW_GUI_NORMAL)

cv2.moveWindow('Main Camera',0,65)

cv2.resizeWindow('Main Camera',W,H)

cv2.namedWindow('Camera1',cv2.WINDOW_GUI_NORMAL)

cv2.moveWindow('Camera1',0,H)

cv2.resizeWindow('Camera1',int(W/4),int(H/4))

cv2.namedWindow('Camera2',cv2.WINDOW_GUI_NORMAL)

cv2.moveWindow('Camera2',int(W/4),H)

cv2.resizeWindow('Camera2',int(W/4),int(H/4))

cv2.namedWindow('Camera3',cv2.WINDOW_GUI_NORMAL)

cv2.moveWindow('Camera3',2*int(W/4),H)

cv2.resizeWindow('Camera3',int(W/4),int(H/4))

cv2.namedWindow('Camera4',cv2.WINDOW_GUI_NORMAL)

cv2.moveWindow('Camera4',3*int(W/4),H)

cv2.resizeWindow('Camera4',int(W/4),int(H/4))

while running:

deltaT = time.time() - tStart

tStart=time.time()

fps = fps*.95 + (1/deltaT)*.05

frame1= piCam1.capture_array()

#frame1=cv2.flip(frame1,-1)

frame2= piCam2.capture_array()

frame2=cv2.flip(frame2,-1)

_ , frame3 = usbCam1.read()

_ , frame4 = usbCam2.read()

mainFrame = frame1

if commandQ.empty() == False:

mainCam = commandQ.get()

if mainCam == 'camera one':

mainFrame = frame1

if mainCam == 'camera two' or mainCam == 'camera to' or mainCam == 'camera too':

mainFrame = frame2

if mainCam == 'camera three':

mainFrame = frame3

if mainCam == 'camera four' or mainCam == 'camera for':

mainFrame = frame4

myText = "FPS: "+str(round(fps,1))

cv2.putText(frame1,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)

cv2.imshow("Main Camera", mainFrame)

cv2.imshow("Camera1", frame1)

cv2.imshow("Camera2", frame2)

cv2.imshow("Camera3", frame3)

cv2.imshow("Camera4", frame4)

if cv2.waitKey(1)==ord('q'):

break

cv2.destroyAllWindows()

print('Program Terminated')

NVIDIA, NVIDIA Jetson Orin Nano

Local Voice Control of NVIDIA Jetson Orin Nano with STT: Getting Started with Vosk

July 22, 2026 admin

Engineering Your Own Local Voice Assistant: No Cloud, No Compromise

Most “smart” voice assistants are just glorified remote controls for someone else’s server. Today, we’re changing that. We are going to build a local, offline voice command pipeline. This isn’t just about saving data; it’s about ownership. When you can control your hardware—like opening and closing a farm gate—without an internet connection, you have built a system that is robust, private, and yours to control forever. Today you are going to get Speech to Text up and running on your NVIDIA Jetson Orin Nano running under Jetpack 7.2.

The “Why” Behind the Setup

You might ask, “Why not just use a cloud API?” Because cloud APIs are fragile. They rely on internet stability, external servers, and privacy-invasive data logging. By running Vosk locally, we keep the processing on your hardware (like the NVIDIA Jetson). It’s faster, it works in the middle of a power-isolated homestead, and it’s 100% secure.

Part 1: Preparing the Environment

Before we can make the machine listen, we have to prepare the battlefield. We aren’t just downloading files; we are setting up a stable environment where your dependencies won’t conflict with your OS.

# Create a dedicated directory so our work stays organized
mkdir -p ~/STT
cd ~/STT

# We need the PortAudio headers. Why? Because Python’s 'pyaudio' is just a wrapper.
# Under the hood, it talks to the C-based PortAudio library. If the headers aren't there,
# the library won't compile, and your microphone will never wake up.
sudo apt update
sudo apt install -y portaudio19-dev wget unzip

# Download the lightweight, low-latency English model weights from Vosk
wget https://alphacephei.com/vosk/models/vosk-model-small-en-us-0.15.zip

# Unzip the archive and rename the folder cleanly to 'model' so the Python script finds it
unzip vosk-model-small-en-us-0.15.zip
mv vosk-model-small-en-us-0.15 model

# Clean up the downloaded zip file to keep the folder pristine
rm vosk-model-small-en-us-0.15.zip

# Setting up a Virtual Environment (venv) is the hallmark of a pro.
# It traps our project dependencies inside this folder so we don't break our system Python.
python3 -m venv sttVenv
source sttVenv/bin/activate
pip install --upgrade pip

# Vosk does the heavy lifting for speech recognition, and Pyaudio grabs the raw data.
pip install vosk pyaudio

# Create a dedicated directory so our work stays organized

mkdir -p ~/STT

cd ~/STT

# We need the PortAudio headers. Why? Because Python’s 'pyaudio' is just a wrapper.

# Under the hood, it talks to the C-based PortAudio library. If the headers aren't there,

# the library won't compile, and your microphone will never wake up.

sudo apt update

sudo apt install -y portaudio19-dev wget unzip

# Download the lightweight, low-latency English model weights from Vosk

wget https://alphacephei.com/vosk/models/vosk-model-small-en-us-0.15.zip

# Unzip the archive and rename the folder cleanly to 'model' so the Python script finds it

unzip vosk-model-small-en-us-0.15.zip

mv vosk-model-small-en-us-0.15 model

# Clean up the downloaded zip file to keep the folder pristine

rm vosk-model-small-en-us-0.15.zip

# Setting up a Virtual Environment (venv) is the hallmark of a pro.

# It traps our project dependencies inside this folder so we don't break our system Python.

python3 -m venv sttVenv

source sttVenv/bin/activate

pip install --upgrade pip

# Vosk does the heavy lifting for speech recognition, and Pyaudio grabs the raw data.

pip install vosk pyaudio

This is IMPORTANT!

Now you can post the code below. You also have to point Thonny to run in the virtual environment. Open Thonny, and under run – select interpreter. Then you must point it to /home/yourUserName/STT/ttsVenv/bin/python3. For me, my username is pjm, but you put in your user name in path above. Here is what mine looked like:

Part 2: Solving the PipeWire Challenge

The biggest headache in modern Linux audio is PipeWire. If you try to open a microphone stream using a hardcoded sample rate that doesn’t match your hardware, your program won’t just fail—it will segfault. We use the validation script below to programmatically query the hardware, asking it: “What sample rate are you running at?” before we even try to open a stream.

# Import the PyAudio library to handle real-time microphone recording and audio stream manipulation.
import pyaudio
# Import the built-in JSON module to parse the structured text results returned by the speech recognition engine.
import json
# Import the audioop library to perform real-time mathematical operations and downsampling on raw audio byte data.
import audioop
# Import the core Vosk objects: Model loads the AI brain/weights, and KaldiRecognizer handles the acoustic processing.
from vosk import Model, KaldiRecognizer
# Import the standard OS module to reliably build absolute system file paths across different user directories.
import os
state=None

# Define a function to discover the specific hardware index and sample rate of the active system sound server.
def get_pipewire_index():
    # Instantiate a PyAudio session object to query the operating system's hardware configuration list.
    p = pyaudio.PyAudio()
    # Loop through every single available audio device reported by the operating system kernel.
    for i in range(p.get_device_count()):
        # Retrieve a descriptive configuration dictionary for the current audio device index.
        dev = p.get_device_info_by_index(i)
        # Check if the name of the device contains 'pipewire' to target the modern Linux audio subsystem.
        if "pipewire" in dev.get('name', '').lower():
            # Return the valid device index integer and its hardware default sample rate, converted cleanly to an integer.
            return i, int(dev.get('defaultSampleRate'))
    # Return empty values if the loop completes without finding an active PipeWire sound server profile.
    return None, None

# Execute the discovery function to capture both the correct system hardware device index and its native sample rate.
dev_index, native_rate = get_pipewire_index()

# Check if the hardware discovery handshake failed to locate an active PipeWire audio interface.
if dev_index is None:
    # Alert the user on screen that the script cannot communicate with the system's active audio server layer.
    print("Error: Could not find 'pipewire' device.")
    # Stop the script execution immediately since it is impossible to record audio without a valid device hook.
    exit()

# Instantiate a fresh PyAudio session specifically dedicated to opening our incoming recording channel.
p = pyaudio.PyAudio()
# Open the active system recording stream channel using explicit audio hardware parameter configurations.
# PARAMETER MANUAL:
# - format=pyaudio.paInt16: Sets 16-bit Signed Integer PCM format. (LEAVE ALONE - required by the Vosk engine).
# - channels=1: Sets recording to Mono channel audio. (LEAVE ALONE - Vosk processes single-channel input).
# - rate=native_rate: Sets the sample rate to match the mic exactly. (LEAVE ALONE - prevents driver hardware failure).
# - input=True: Configures this stream as an audio capture device. (LEAVE ALONE - required to wake up the microphone).
# - input_device_index=dev_index: Points to our discovered target device. (LEAVE ALONE - ensures we grab the right mic).
# - frames_per_buffer=4096: Controls audio chunk size. (PLAY WITH: Lower values like 2048 reduce latency but risk buffer overflows; higher values like 8192 increase safety but introduce slight latency).
stream = p.open(format=pyaudio.paInt16, channels=1, rate=native_rate, 
                input=True, input_device_index=dev_index, frames_per_buffer=4096)

# Generate an absolute directory path targeting the exact folder containing your extracted acoustic model weights.
# PARAMETER MANUAL:
# - "STT", "model": Represents folder names. (PLAY WITH: Change these only if your physical folder layout uses different names).
model_path = os.path.join(os.path.expanduser("~"), "STT", "model")
# Load the heavy machine learning weights into system memory to initialize the Vosk voice processing brain.
model = Model(model_path)
# Initialize the primary speech recognizer object using the loaded model weights and target processing frequency.
# PARAMETER MANUAL:
# - model: The initialized AI brain variable. (LEAVE ALONE).
# - 16000: Specifies the target processing sample rate in Hertz. (LEAVE ALONE - Vosk models are mathematically trained for exactly 16000Hz).
recognizer = KaldiRecognizer(model, 16000)

# Print a status message to the terminal showing that the handshake succeeded and revealing the mic's operating frequency.
print(f"Listening on PipeWire at {native_rate}Hz...")
# Turn on the hardware recording valve to begin filling up the system input data buffers.
stream.start_stream()

# Set up an exception safety block to intercept manual user exits gracefully without leaving system resources locked up.
try:
    # Launch the infinite real-time audio capture loop to continuously monitor the microphone stream.
    while True:
        # Pull raw binary audio frame bytes directly out of the incoming hardware stream buffer memory block.
        # PARAMETER MANUAL:
        # - 4096: The chunk size to read. (LEAVE ALONE - must perfectly match your stream's frames_per_buffer setting).
        # - exception_on_overflow=False: Prevents Python crashes if the Orin momentarily falls behind. (LEAVE ALONE).
        data = stream.read(4096, exception_on_overflow=False)
        # Downsample the raw audio bytes mathematically to 16000Hz using digital filtering algorithms.
        # PARAMETER MANUAL:
        # - data: Raw input audio byte string. (LEAVE ALONE).
        # - 2: Size of sample width in bytes for 16-bit sound. (LEAVE ALONE).
        # - 1: Input audio channel count. (LEAVE ALONE).
        # - native_rate: The native frequency of your mic. (LEAVE ALONE).
        # - 16000: The required downsampled output frequency. (LEAVE ALONE).
        # - state: The digital filtering conversion state history object. (LEAVE ALONE).
        # CRITICAL NOTE: The 'state' variable must be initialized as 'state = None' before entering this 'try:' block, 
        # otherwise Python will throw a NameError on the very first loop iteration because 'state' does not exist yet.
        data_16k, state = audioop.ratecv(data, 2, 1, native_rate, 16000, state)
        
        # Feed the processed 16kHz audio data block into Vosk to evaluate if a complete spoken phrase has finished.
        if recognizer.AcceptWaveform(data_16k):
            # Parse the structured JSON text string returned by the acoustic model into a standard Python dictionary.
            res = json.loads(recognizer.Result())
            # Check the parsed result dictionary to ensure that the detected text sequence is not completely blank.
            if res.get("text"):
                # Cleanly output the decoded text command straight to the standard system terminal display screen.
                print("Command: "+res['text'])
# Intercept the user pressing Ctrl+C in the terminal to stop the continuous script loop cleanly.
except KeyboardInterrupt:
    # Print a clean closing status notification on a new line to confirm the signal was intercepted.
    print("\nStopped.")
# Execute mandatory clean up instructions that are guaranteed to run no matter how the script terminates or crashes.
finally:
    # Turn off the active recording stream data valve to stop pulling raw information from the audio driver layer.
    stream.stop_stream()
    # Close the recording stream channel completely to drop the active hardware pointer binding handles.
    stream.close()
    # Terminate the parent PyAudio engine instance to release all allocated audio device resources back to the OS.
    p.terminate()

100

101

102

103

104

105

106

107

108

# Import the PyAudio library to handle real-time microphone recording and audio stream manipulation.

import pyaudio

# Import the built-in JSON module to parse the structured text results returned by the speech recognition engine.

import json

# Import the audioop library to perform real-time mathematical operations and downsampling on raw audio byte data.

import audioop

# Import the core Vosk objects: Model loads the AI brain/weights, and KaldiRecognizer handles the acoustic processing.

from vosk import Model, KaldiRecognizer

# Import the standard OS module to reliably build absolute system file paths across different user directories.

import os

state=None

# Define a function to discover the specific hardware index and sample rate of the active system sound server.

def get_pipewire_index():

# Instantiate a PyAudio session object to query the operating system's hardware configuration list.

p = pyaudio.PyAudio()

# Loop through every single available audio device reported by the operating system kernel.

for i in range(p.get_device_count()):

# Retrieve a descriptive configuration dictionary for the current audio device index.

dev = p.get_device_info_by_index(i)

# Check if the name of the device contains 'pipewire' to target the modern Linux audio subsystem.

if "pipewire" in dev.get('name', '').lower():

# Return the valid device index integer and its hardware default sample rate, converted cleanly to an integer.

return i, int(dev.get('defaultSampleRate'))

# Return empty values if the loop completes without finding an active PipeWire sound server profile.

return None, None

# Execute the discovery function to capture both the correct system hardware device index and its native sample rate.

dev_index, native_rate = get_pipewire_index()

# Check if the hardware discovery handshake failed to locate an active PipeWire audio interface.

if dev_index is None:

# Alert the user on screen that the script cannot communicate with the system's active audio server layer.

print("Error: Could not find 'pipewire' device.")

# Stop the script execution immediately since it is impossible to record audio without a valid device hook.

exit()

# Instantiate a fresh PyAudio session specifically dedicated to opening our incoming recording channel.

p = pyaudio.PyAudio()

# Open the active system recording stream channel using explicit audio hardware parameter configurations.

# PARAMETER MANUAL:

# - format=pyaudio.paInt16: Sets 16-bit Signed Integer PCM format. (LEAVE ALONE - required by the Vosk engine).

# - channels=1: Sets recording to Mono channel audio. (LEAVE ALONE - Vosk processes single-channel input).

# - rate=native_rate: Sets the sample rate to match the mic exactly. (LEAVE ALONE - prevents driver hardware failure).

# - input=True: Configures this stream as an audio capture device. (LEAVE ALONE - required to wake up the microphone).

# - input_device_index=dev_index: Points to our discovered target device. (LEAVE ALONE - ensures we grab the right mic).

# - frames_per_buffer=4096: Controls audio chunk size. (PLAY WITH: Lower values like 2048 reduce latency but risk buffer overflows; higher values like 8192 increase safety but introduce slight latency).

stream = p.open(format=pyaudio.paInt16, channels=1, rate=native_rate,

input=True, input_device_index=dev_index, frames_per_buffer=4096)

# Generate an absolute directory path targeting the exact folder containing your extracted acoustic model weights.

# PARAMETER MANUAL:

# - "STT", "model": Represents folder names. (PLAY WITH: Change these only if your physical folder layout uses different names).

model_path = os.path.join(os.path.expanduser("~"), "STT", "model")

# Load the heavy machine learning weights into system memory to initialize the Vosk voice processing brain.

model = Model(model_path)

# Initialize the primary speech recognizer object using the loaded model weights and target processing frequency.

# PARAMETER MANUAL:

# - model: The initialized AI brain variable. (LEAVE ALONE).

# - 16000: Specifies the target processing sample rate in Hertz. (LEAVE ALONE - Vosk models are mathematically trained for exactly 16000Hz).

recognizer = KaldiRecognizer(model, 16000)

# Print a status message to the terminal showing that the handshake succeeded and revealing the mic's operating frequency.

print(f"Listening on PipeWire at {native_rate}Hz...")

# Turn on the hardware recording valve to begin filling up the system input data buffers.

stream.start_stream()

# Set up an exception safety block to intercept manual user exits gracefully without leaving system resources locked up.

try:

# Launch the infinite real-time audio capture loop to continuously monitor the microphone stream.

while True:

# Pull raw binary audio frame bytes directly out of the incoming hardware stream buffer memory block.

# PARAMETER MANUAL:

# - 4096: The chunk size to read. (LEAVE ALONE - must perfectly match your stream's frames_per_buffer setting).

# - exception_on_overflow=False: Prevents Python crashes if the Orin momentarily falls behind. (LEAVE ALONE).

data = stream.read(4096, exception_on_overflow=False)

# Downsample the raw audio bytes mathematically to 16000Hz using digital filtering algorithms.

# PARAMETER MANUAL:

# - data: Raw input audio byte string. (LEAVE ALONE).

# - 2: Size of sample width in bytes for 16-bit sound. (LEAVE ALONE).

# - 1: Input audio channel count. (LEAVE ALONE).

# - native_rate: The native frequency of your mic. (LEAVE ALONE).

# - 16000: The required downsampled output frequency. (LEAVE ALONE).

# - state: The digital filtering conversion state history object. (LEAVE ALONE).

# CRITICAL NOTE: The 'state' variable must be initialized as 'state = None' before entering this 'try:' block,

# otherwise Python will throw a NameError on the very first loop iteration because 'state' does not exist yet.

data_16k, state = audioop.ratecv(data, 2, 1, native_rate, 16000, state)

# Feed the processed 16kHz audio data block into Vosk to evaluate if a complete spoken phrase has finished.

if recognizer.AcceptWaveform(data_16k):

# Parse the structured JSON text string returned by the acoustic model into a standard Python dictionary.

res = json.loads(recognizer.Result())

# Check the parsed result dictionary to ensure that the detected text sequence is not completely blank.

if res.get("text"):

# Cleanly output the decoded text command straight to the standard system terminal display screen.

print("Command: "+res['text'])

# Intercept the user pressing Ctrl+C in the terminal to stop the continuous script loop cleanly.

except KeyboardInterrupt:

# Print a clean closing status notification on a new line to confirm the signal was intercepted.

print("\nStopped.")

# Execute mandatory clean up instructions that are guaranteed to run no matter how the script terminates or crashes.

finally:

# Turn off the active recording stream data valve to stop pulling raw information from the audio driver layer.

stream.stop_stream()

# Close the recording stream channel completely to drop the active hardware pointer binding handles.

stream.close()

# Terminate the parent PyAudio engine instance to release all allocated audio device resources back to the OS.

p.terminate()

Homework: Your Gate Controller

You now have a system that identifies audio input, resamples it to 16kHz, and outputs text. Your assignment: Transform this text output into an action.

I want you to add a conditional statement to the main loop. If the recognized text is “open”, print an ASCII art representation of an open gate. If it’s “close”, print the closed version. This is the first step in closing the loop between your AI and the physical world. Go get ’em, and don’t just copy the code—understand how the data flows from the microphone to your decision logic!

AI On the Edge, Raspberry Pi

AI on the Edge LESSON 34: Project Combining TTS, STT, Face Recognition and Servos on Pi 5

July 18, 2026 admin

In our previous lessons, we built individual components. We got our machine to listen locally using Speech-to-Text (STT), we got it to talk back cleanly with Text-to-Speech (TTS), and we learned how to manipulate physical hardware using precision servos. Today, we bridge the gap between software and the physical world. We are building a multi-threaded, autonomous edge system that tracks humanoids in real time.

The Engineering Mindset: Concurrency & Threading

If you try to build a system like this sequentially—running camera capture, object detection, audio listening, and audio speaking in a single while True: loop—your project will fail completely. Why? Because tasks like waiting for a voice command (stt.listen()) or synthesizing a voice line are blocking operations. If your code is stuck waiting for you to finish speaking a command, your camera frames freeze, your servo adjustments drop to zero, and your humanoid target escapes.

To solve this, we architect our software using Multi-threading. We spawn independent execution paths that run concurrently, passing data back and forth safely using thread-safe Queues:

The Main Thread: Handles the high-speed Picamera2 video capture loop, executes Haar Cascade face/eye calculations, and updates the servo angles via mathematical error tracking.
The Speak Thread: Idles quietly in the background until a message lands in the speakQ, instantly triggering the local Piper TTS engine without stalling our camera frame rate.
The Command Thread: Keeps a continuous ear open via our local STT engine. When it parses a voice command like “track”, “release”, or “quit”, it safely pops that token into the commandQ for the main loop to execute on its next pass.

The Complete Fusion Architecture Code

Below is the complete Python pipeline developed in today’s lesson. Make sure your hardware connections for the pan-and-tilt servos match pins 2 and 3 on your expansion setup, and verify your fusion_hat software stack is completely updated.

import cv2
import time
from picamera2 import Picamera2
#########################################
import threading
from queue import Queue
from fusion_hat.tts import Piper
tts = Piper()
tts.set_model('en_US-kristin-medium')

from fusion_hat.stt import STT
stt = STT('en-us')

cnt = 0
running = True
commandQ = Queue()
speakQ = Queue()
###########################################

from fusion_hat.servo import Servo
panPin = 2
tiltPin = 3
panServo = Servo(panPin)
tiltServo = Servo(tiltPin)
panAngle = -10
tiltAngle =-20
panServo.angle(panAngle)
tiltServo.angle(tiltAngle)
piCam = Picamera2()
W=1280
H=720
tStart = time.time()
fps = 0
RES = (W,H)
piCam.preview_configuration.main.size = RES
piCam.preview_configuration.main.format = "RGB888"
piCam.preview_configuration.controls.FrameRate=60
piCam.preview_configuration.align()
piCam.configure("preview")
piCam.start()

textLowerLeft = (int(W*.01),int(H*.05))
fontFace = cv2.FONT_HERSHEY_SIMPLEX
fontThickness = int(W/425)
fontScale = H*.0015
fontColor = (0,0,255)

faceFinder = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
eyeFinder = cv2.CascadeClassifier(cv2.data.haarcascades + "haarcascade_eye.xml")

track = False
xFrameC = int((W-1)/2)
yFrameC = int((H-1)/2)
xBoxC = xFrameC
yBoxC = yFrameC
##########################################
def speak():
    global running
    print('Speak Thread Running')
    while running:
        if speakQ.empty() == False:
            txt = speakQ.get()
            print(txt)
            tts.say(txt)
    print('Speak Thread is Terminated')
    print('')
def getCommand():
    print('Get Command Thread is Running')
    global running
    while running:
        myCommand = stt.listen(stream = False)
        myCommand = myCommand.strip()
        print(myCommand)
        if myCommand == 'quit':
            msg = 'Project Humanoid Elimination Terminated. Have a Nice Day.'
            speakQ.put(msg)
            running = False
            break
        commandQ.put(myCommand)
    print('Get Command Thread Terminated')
    print('')

speakThread = threading.Thread(target=speak,daemon=True)
speakThread.start()

commandThread = threading.Thread(target=getCommand,daemon=True)
commandThread.start()
print('Main Program Has Commenced')
msg = 'Hello, I am Ready to Eliminate Humanoids'
speakQ.put(msg)
msgOld = 'Hello, I am Ready to Eliminate Humanoids'

cv2.namedWindow('Camera',cv2.WINDOW_GUI_NORMAL)
cv2.moveWindow('Camera',0,65)
cv2.resizeWindow('Camera',W,H)
############################################
while running:
    deltaT = time.time() - tStart
    tStart=time.time()
    fps = fps*.95 + (1/deltaT)*.05
    frame= piCam.capture_array()
    frame=cv2.flip(frame,-1)
    
    if commandQ.empty() == False:
        myCommand = commandQ.get()
        if myCommand == 'track':
            track = True
        if myCommand == 'release':
            track = False

    gray = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)
    faces = faceFinder.detectMultiScale(gray,scaleFactor=1.1, minNeighbors =5, minSize=(100,100))
    if len(faces) != 0:
        cnt = 0 
        msg = "Humanoid Detected, Shall I track?"
    if len(faces) == 0:
        cnt = cnt + 1
        if cnt == 25:
            msg = "No Humanoid Detected, Continuing to Scan"
    if msgOld != msg:
        speakQ.put(msg)
    msgOld = msg
    for (x,y,w,h) in faces:
        cv2.rectangle(frame, (x,y),(x+w,y+h),(0,255,255),3)
        
        xBoxC = x + int(w/2)
        yBoxC = y + int(h/2)
        xError = xBoxC - xFrameC
        yError = yBoxC - yFrameC
        
        eyes =eyeFinder.detectMultiScale(gray[y:y+h,x:x+w],scaleFactor=1.1,minNeighbors=5,minSize=(15,15))
        for (i,j,w,h) in eyes:
            cv2.rectangle(frame,(x+i,y+j),(x+i+w,y+j+h),(255,0,0),3)
    
    if track == True and len(faces)&gt;0:
        panAngle = panAngle - xError/50/2
        panServo.angle(int(panAngle))
        #time.sleep(.02)
        tiltAngle = tiltAngle + yError/50/2
        tiltServo.angle(int(tiltAngle))
        #time.sleep(.02)
                        
    
    myText = "FPS: "+str(round(fps,1))
    cv2.putText(frame,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)
    cv2.imshow("Camera", frame)
    cv2.moveWindow("Camera",0,60)
    if cv2.waitKey(1)==ord('q'):
        break
cv2.destroyAllWindows()
print('')
print('Program Terminated')

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

import cv2

import time

from picamera2 import Picamera2

#########################################

import threading

from queue import Queue

from fusion_hat.tts import Piper

tts = Piper()

tts.set_model('en_US-kristin-medium')

from fusion_hat.stt import STT

stt = STT('en-us')

cnt = 0

running = True

commandQ = Queue()

speakQ = Queue()

###########################################

from fusion_hat.servo import Servo

panPin = 2

tiltPin = 3

panServo = Servo(panPin)

tiltServo = Servo(tiltPin)

panAngle = -10

tiltAngle =-20

panServo.angle(panAngle)

tiltServo.angle(tiltAngle)

piCam = Picamera2()

W=1280

H=720

tStart = time.time()

fps = 0

RES = (W,H)

piCam.preview_configuration.main.size = RES

piCam.preview_configuration.main.format = "RGB888"

piCam.preview_configuration.controls.FrameRate=60

piCam.preview_configuration.align()

piCam.configure("preview")

piCam.start()

textLowerLeft = (int(W*.01),int(H*.05))

fontFace = cv2.FONT_HERSHEY_SIMPLEX

fontThickness = int(W/425)

fontScale = H*.0015

fontColor = (0,0,255)

faceFinder = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')

eyeFinder = cv2.CascadeClassifier(cv2.data.haarcascades + "haarcascade_eye.xml")

track = False

xFrameC = int((W-1)/2)

yFrameC = int((H-1)/2)

xBoxC = xFrameC

yBoxC = yFrameC

##########################################

def speak():

global running

print('Speak Thread Running')

while running:

if speakQ.empty() == False:

txt = speakQ.get()

print(txt)

tts.say(txt)

print('Speak Thread is Terminated')

print('')

def getCommand():

print('Get Command Thread is Running')

global running

while running:

myCommand = stt.listen(stream = False)

myCommand = myCommand.strip()

print(myCommand)

if myCommand == 'quit':

msg = 'Project Humanoid Elimination Terminated. Have a Nice Day.'

speakQ.put(msg)

running = False

break

commandQ.put(myCommand)

print('Get Command Thread Terminated')

print('')

speakThread = threading.Thread(target=speak,daemon=True)

speakThread.start()

commandThread = threading.Thread(target=getCommand,daemon=True)

commandThread.start()

print('Main Program Has Commenced')

msg = 'Hello, I am Ready to Eliminate Humanoids'

speakQ.put(msg)

msgOld = 'Hello, I am Ready to Eliminate Humanoids'

cv2.namedWindow('Camera',cv2.WINDOW_GUI_NORMAL)

cv2.moveWindow('Camera',0,65)

cv2.resizeWindow('Camera',W,H)

############################################

while running:

deltaT = time.time() - tStart

tStart=time.time()

fps = fps*.95 + (1/deltaT)*.05

frame= piCam.capture_array()

frame=cv2.flip(frame,-1)

if commandQ.empty() == False:

myCommand = commandQ.get()

if myCommand == 'track':

track = True

if myCommand == 'release':

track = False

gray = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)

faces = faceFinder.detectMultiScale(gray,scaleFactor=1.1, minNeighbors =5, minSize=(100,100))

if len(faces) != 0:

cnt = 0

msg = "Humanoid Detected, Shall I track?"

if len(faces) == 0:

cnt = cnt + 1

if cnt == 25:

msg = "No Humanoid Detected, Continuing to Scan"

if msgOld != msg:

speakQ.put(msg)

msgOld = msg

for (x,y,w,h) in faces:

cv2.rectangle(frame, (x,y),(x+w,y+h),(0,255,255),3)

xBoxC = x + int(w/2)

yBoxC = y + int(h/2)

xError = xBoxC - xFrameC

yError = yBoxC - yFrameC

eyes =eyeFinder.detectMultiScale(gray[y:y+h,x:x+w],scaleFactor=1.1,minNeighbors=5,minSize=(15,15))

for (i,j,w,h) in eyes:

cv2.rectangle(frame,(x+i,y+j),(x+i+w,y+j+h),(255,0,0),3)

if track == True and len(faces)>0:

panAngle = panAngle - xError/50/2

panServo.angle(int(panAngle))

#time.sleep(.02)

tiltAngle = tiltAngle + yError/50/2

tiltServo.angle(int(tiltAngle))

#time.sleep(.02)

myText = "FPS: "+str(round(fps,1))

cv2.putText(frame,myText,textLowerLeft,fontFace,fontScale,fontColor,fontThickness)

cv2.imshow("Camera", frame)

cv2.moveWindow("Camera",0,60)

if cv2.waitKey(1)==ord('q'):

break

cv2.destroyAllWindows()

print('')

print('Program Terminated')

Breaking Down the Math & Logic

Pro-Tip on State Debouncing: Notice how we handle messages using the cnt variable and the msgOld buffer. If we simply asked the system to say “Humanoid Detected” every single time the frame loops, the TTS engine would crash from queue overflow, stuttering continuously. By checking if msgOld != msg, we ensure the system speaks a notification exactly once upon transition. The 25-frame hysteresis buffer ensures that a single missed frame doesn’t cause a false “No Humanoid” trigger.

Proportional Error Tracking

Look carefully at how the servo angles are adjusted:

panAngle = panAngle - xError/50/2

1	panAngle = panAngle - xError/50/2

We are calculating the error vector—how many pixels the center of the bounding box (xBoxC) is from the absolute center of our camera stream (xFrameC). Instead of moving the servo by a fixed step, we move it proportionally to the size of the error. Big error? Fast movement. Small error? Gentle, microscopic correction. Dividing by 50 dampens our proportional gain loop so our servos don’t violently oscillate and throw our camera system completely out of alignment.

Your Homework Assignment

You have the system reading frames, listening asynchronously, and adjusting angles proportionally. Now it’s time to earn your stripes as an edge systems designer.

Your Assignment: Look at our msg tracking logic. Right now, if a humanoid is detected, it simply asks, “Shall I track?”. Your task is to tie our speech input directly to this state. Modify the control flow so that the tracking logic requires verification. If a humanoid is detected, the system must wait until the commandQ yields a verified verbal confirmation—like “execute” or “yes”—before the proportional servo tracking begins. If it hears “abort”, it must break off and look away.

Drop your custom implementation script in the comments section below, explain the logic behind your state modifications, and let’s see how you optimize your loop efficiency. Go get ’em!

AI On the Edge, Tutorial

AI on the Edge LESSON 13: Control LED Brightness with Voice Commands on Raspberry Pi 5

May 9, 2026 admin

Hey everyone, welcome back to the AI on the Edge series!

In today’s lesson, we’re taking another big step forward in building truly interactive AI projects that run right on our Raspberry Pi 5. We’re going to give our hardware a voice — literally. You’ll learn how to control the brightness of an LED using simple voice commands like “low”, “medium”, “high”, “on”, and “off”.

This lesson builds directly on the speech-to-text skills we learned earlier. Using the Fusion Hat’s microphone and the excellent STT library, we create a system where you can speak naturally to your Pi and it responds instantly by changing the LED brightness. We also bring in Python threading so the voice listening doesn’t block the main program — which is a critical skill as our projects get more complex.

One of the things I really like about this project is how it shows the power of combining AI with real hardware. You’re not just making the LED turn on and off anymore — you’re giving it smooth, adjustable brightness control using nothing but your voice. It’s a perfect example of the kind of interactive, intelligent edge computing we’re working toward in this class.

By the end of this lesson, you’ll have a solid understanding of how to use voice commands to control hardware, how to manage multiple things happening at the same time with threading, and how to create a much more natural and user-friendly interface for your projects.

This is the kind of thing that makes your Raspberry Pi projects feel alive and responsive. Whether you eventually want to control motors, lights, robots, or entire systems with your voice, the techniques you learn in this lesson will serve as a strong foundation.

So grab your SunFounder Fusion AI Hat, hook up that red LED, and let’s get your Raspberry Pi listening and responding to your voice commands like a proper smart device!

As always, I encourage you to type the code along with me in the video, then play around with it. Try adding more commands, change the LED color, or combine it with other sensors. That’s where the real learning and creativity happens.

I’m really excited to see what you build with this one!

This is the schematic we are using, from LESSON #5.

Fusion Hat Circuit Diagram — This is the circuit we will use moving forward in the class

In the video, this is the code we developed:

from fusion_hat.pwm import PWM
from fusion_hat.stt import STT
import threading
from queue import Queue
from time import sleep

redPin = 5
redLED = PWM(redPin)
redLED.freq(1000)

stt = STT(language ="en-us")

brightness = "off"
bNumb = 0
running = True
brightQ = Queue()
def blinkBright():
    print("Input Thread is Running")
    global running
    while running:
        print("What Brightness: (off, low, medium, high, on or quit: )")
        cmd = stt.listen(stream = False)
        cmd = cmd.strip()
        print(cmd)
        if cmd == "quit":
            running = False
            break
        brightQ.put(cmd)
    print("Thread is Terminated")
brightThread = threading.Thread(target = blinkBright, daemon = True)
brightThread.start()
print("Main Program Has Started")
try:
    while running:
        if brightQ.empty() == False:
            brightness = brightQ.get()
            if brightness == "quit":
                break
        if brightness == "off":
            bNumb = 0
        if brightness == "low" or brightness == "whoa":
            bNumb = 8
        if brightness == "medium":
            bNumb = 25
        if brightness == "high" or brightness == "hi":
            bNumb = 60
        if brightness == "on":
            bNumb = 100
        redLED.pulse_width_percent(bNumb)
    redLED.pulse_width_percent(0)
    redLED.enable(False)
    print("LED Released")
    print("Program is Terminated")
        
except KeyboardInterrupt:
    running = False
    redLED.pulse_width_percent(0)
    redLED.enable(False)
    print("LED Released")
    print("Program Terminated")

from fusion_hat.pwm import PWM

from fusion_hat.stt import STT

import threading

from queue import Queue

from time import sleep

redPin = 5

redLED = PWM(redPin)

redLED.freq(1000)

stt = STT(language ="en-us")

brightness = "off"

bNumb = 0

running = True

brightQ = Queue()

def blinkBright():

print("Input Thread is Running")

global running

while running:

print("What Brightness: (off, low, medium, high, on or quit: )")

cmd = stt.listen(stream = False)

cmd = cmd.strip()

print(cmd)

if cmd == "quit":

running = False

break

brightQ.put(cmd)

print("Thread is Terminated")

brightThread = threading.Thread(target = blinkBright, daemon = True)

brightThread.start()

print("Main Program Has Started")

try:

while running:

if brightQ.empty() == False:

brightness = brightQ.get()

if brightness == "quit":

break

if brightness == "off":

bNumb = 0

if brightness == "low" or brightness == "whoa":

bNumb = 8

if brightness == "medium":

bNumb = 25

if brightness == "high" or brightness == "hi":

bNumb = 60

if brightness == "on":

bNumb = 100

redLED.pulse_width_percent(bNumb)

redLED.pulse_width_percent(0)

redLED.enable(False)

print("LED Released")

print("Program is Terminated")

except KeyboardInterrupt:

running = False

redLED.pulse_width_percent(0)

redLED.enable(False)

print("LED Released")

print("Program Terminated")

AI On the Edge, Tutorial

AI on the Edge LESSON 11: Control LED on Raspberry Pi With Voice Commands

May 1, 2026 admin

In today’s lesson, we’re taking our first exciting step into giving our Raspberry Pi the ability to understand and respond to our voice. That’s right — we’re going to control a physical LED using nothing but spoken commands! This is a huge milestone in the class because it marks the beginning of building truly interactive AI projects that can listen to us and take action in the real world.

Using the SunFounder Fusion AI Hat’s built-in microphone and the excellent STT (Speech-to-Text) library, we create a simple but powerful voice assistant that can turn an LED on and off with commands like “on”, “off”, and “quit”. I walk you through every single line of the code so you can clearly see how we capture voice input, process the command, and control real hardware.

This lesson is intentionally straightforward because I want you to build a strong foundation. Once you understand how to take a voice command and turn it into physical action, we can start adding more complexity — like controlling multiple devices, adjusting brightness, or even combining voice control with computer vision in future lessons.

One of the things I love most about this project is how it makes the Raspberry Pi feel “alive.” Instead of clicking buttons or typing commands, you can now talk directly to your project. This is the kind of interaction that makes edge AI projects so much fun and so powerful.

By the end of this lesson, you’ll have a working voice-controlled LED and the confidence to start expanding your voice control skills. This is exactly the kind of capability we need as we move forward in the AI on the Edge journey — giving our intelligent systems natural, human-friendly ways to interact with us.

So grab your Fusion AI Hat, hook up that LED, and let’s turn your Raspberry Pi into a voice-controlled device! As always, I strongly encourage you to code along with me in the video and then play around with the program. Try adding more commands, control multiple LEDs, or even have it say something back to you.

This is where things start getting really fun. Let’s get that LED responding to your voice!

This is the schematic of the circuit we are using for our AI class. We go into great detail on this schematic in LESSON #5 if you want to learn more about it.

Now this is the code we developed in this lesson:

from fusion_hat.stt import STT
from fusion_hat.pin import Pin, Mode
from time import sleep

redPin = 17
redLED = Pin(redPin,mode=Mode.OUT)
stt = STT(language="en-us")
while True:
    print("Voice Assistant At the READY!, Speak at Any Time: ")
    command = stt.listen(stream=False)
    command = command.strip()
    print(command)
    if command == "on":
        redLED.high()
        print("LED ON")
    if command == "off":
        redLED.low()
        print("LED OFF")
    if command == "quit":
        break
redLED.low()

from fusion_hat.stt import STT

from fusion_hat.pin import Pin, Mode

from time import sleep

redPin = 17

redLED = Pin(redPin,mode=Mode.OUT)

stt = STT(language="en-us")

while True:

print("Voice Assistant At the READY!, Speak at Any Time: ")

command = stt.listen(stream=False)

command = command.strip()

print(command)

if command == "on":

redLED.high()

print("LED ON")

if command == "off":

redLED.low()

print("LED OFF")

if command == "quit":

break

redLED.low()

Technology Tutorials

Tag Archives: Speech to Text

AI on the Edge LESSON 36: Select Active Camera in OpenCV With Voice Commands

What You Will Learn in This Lesson

Step-by-Step Breakdown of the Script

1. Setting Up the Background Voice Thread

2. Initializing Multiple Cameras

3. Processing and Routing Commands in the Main Loop

4. Managing Your OpenCV Windows

Local Voice Control of NVIDIA Jetson Orin Nano with STT: Getting Started with Vosk

Engineering Your Own Local Voice Assistant: No Cloud, No Compromise

The “Why” Behind the Setup

Part 1: Preparing the Environment

This is IMPORTANT!

Part 2: Solving the PipeWire Challenge

Homework: Your Gate Controller

AI on the Edge LESSON 34: Project Combining TTS, STT, Face Recognition and Servos on Pi 5

The Engineering Mindset: Concurrency & Threading

The Complete Fusion Architecture Code

Breaking Down the Math & Logic

Proportional Error Tracking

Your Homework Assignment

AI on the Edge LESSON 13: Control LED Brightness with Voice Commands on Raspberry Pi 5

AI on the Edge LESSON 11: Control LED on Raspberry Pi With Voice Commands

Making The World a Better Place One High Tech Project at a Time. Enjoy!