All posts by admin

In this video lesson you will learn how to train the Raspberry Pi to take voice commands from you. We do this through the Fusion AI+ hat’s microphone, and a Speech to Text (STT) model. Our goal is to develop the ability to interact with our projects through spoken words. We give commands to the project, and then it responds intelligently with words.

Remember these lessons depend on you using the AI Educational OS, a special version of bookworm that has all the libraries, modules, and models already installed for you. See LESSON #2 in this class for instructions on flashing that OS.

Below is the simple demonstration code we developed to give simple voice commands:

from fusion_hat.stt import STT

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

while True:
    print("Say Something: ")
    command = stt.listen(stream=False)
    print(command)

from fusion_hat.stt import STT

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

while True:

print("Say Something: ")

command = stt.listen(stream=False)

print(command)

Similar to our Speech to Text example, the first time you run this program you will get a permissions error. You need to open a terminal, and put these commands in one at a time to enable permissions. This only has to be done once, and after that this and all STT programs should run properly.

sudo mkdir -p /opt/vosk_models
sudo chown -R pi:pi /opt/vosk_models
sudo chmod -R 755 /opt/vosk_models

sudo mkdir -p /opt/vosk_models

sudo chown -R pi:pi /opt/vosk_models

sudo chmod -R 755 /opt/vosk_models

AI On the Edge

AI on the Edge LESSON 9: Make Your Raspberry Pi Talk with TTS

April 26, 2026 admin

In this video lesson I show you how to make your Pi Project talk to you using Text to Speech (TTS). We will use the Piper Model which gives good clear voice, and a choice from a large number of different models.

In this lesson I show you my solution to the Homework assigned in LESSON #8.

In this lesson we demonstrate this program, which makes the Pi talk, through the Fusion HAT+ speaker. The homework was to create a program that smoothly adjust the brightness of an LED using a potentiometer. The project was also to speak to the user, indicating what brightness the LED was set at.

This is the schematic we are using on these projects;

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

This Schematic is explained in detail in LESSON #5.

Then this is the code we developed in today’s lesson.

from fusion_hat.adc import ADC
from fusion_hat.pwm import PWM
from fusion_hat.tts import Piper
from time import sleep
tts = Piper()
tts.set_model('en_US-amy-low')
potPin = 0
myPot = ADC(potPin)
redPin = 5
redLED = PWM(redPin)
rangeOld = 100

while True:
    potVal = myPot.read()
    writeVal=100*(potVal/4095) - .005
    if writeVal < 1:
        msg = "Your LED is Off"
        range = 0
    if writeVal >=1 and writeVal<10:
        msg = "Your LED is Brightness 1"
        range =  1
    if writeVal >=10 and writeVal<20:
        msg = "Your LED is Brightness 2"
        range=2
    if writeVal >=20 and writeVal<30:
        msg = "Your LED is Brightness 3"
        range=3
    if writeVal >=30 and writeVal<40:
        msg = "Your LED is Brightness 4"
        range=4
    if writeVal >=40 and writeVal<50:
        msg = "Your LED is Brightness 5"
        range=5
    if writeVal >=50 and writeVal<60:
        msg = "Your LED is Brightness 6"
        range=6
    if writeVal >=60 and writeVal<70:
        msg = "Your LED is Brightness 7"
        range=7
    if writeVal >=70 and writeVal<80:
        msg = "Your LED is Brightness 8"
        range=8
    if writeVal >=80 and writeVal<90:
        msg = "Your LED is Brightness 9"
        range=9
    if writeVal >=90 and writeVal<99:
        msg = "Your LED is Brightness 10"
        range=10
    if writeVal >=99:
        msg = "Your LED is full Brightness"
        range=11
    if rangeOld != range:
        tts.say(msg, stream = False)
    rangeOld=range
    redLED.pulse_width_percent(int(writeVal))
    sleep(.1)

from fusion_hat.adc import ADC

from fusion_hat.pwm import PWM

from fusion_hat.tts import Piper

from time import sleep

tts = Piper()

tts.set_model('en_US-amy-low')

potPin = 0

myPot = ADC(potPin)

redPin = 5

redLED = PWM(redPin)

rangeOld = 100

while True:

potVal = myPot.read()

writeVal=100*(potVal/4095) - .005

if writeVal < 1:

msg = "Your LED is Off"

range = 0

if writeVal >=1 and writeVal<10:

msg = "Your LED is Brightness 1"

range = 1

if writeVal >=10 and writeVal<20:

msg = "Your LED is Brightness 2"

range=2

if writeVal >=20 and writeVal<30:

msg = "Your LED is Brightness 3"

range=3

if writeVal >=30 and writeVal<40:

msg = "Your LED is Brightness 4"

range=4

if writeVal >=40 and writeVal<50:

msg = "Your LED is Brightness 5"

range=5

if writeVal >=50 and writeVal<60:

msg = "Your LED is Brightness 6"

range=6

if writeVal >=60 and writeVal<70:

msg = "Your LED is Brightness 7"

range=7

if writeVal >=70 and writeVal<80:

msg = "Your LED is Brightness 8"

range=8

if writeVal >=80 and writeVal<90:

msg = "Your LED is Brightness 9"

range=9

if writeVal >=90 and writeVal<99:

msg = "Your LED is Brightness 10"

range=10

if writeVal >=99:

msg = "Your LED is full Brightness"

range=11

if rangeOld != range:

tts.say(msg, stream = False)

rangeOld=range

redLED.pulse_width_percent(int(writeVal))

sleep(.1)

AI On the Edge, Tutorial

AI on the Edge LESSON 8: Text to Speech (TTS) on the Raspberry Pi

April 24, 2026 admin

In this video lesson I will show you how to get the Raspberry Pi to speak to you in plain English. This is our first dabbling with AI. In earlier lessons we have discussed that one of our first objectives will be to begin to audibly interact with our project through speech. The first step will be to get the Pi to talk to us. Then in future lessons we will show how to get the Pi to listen to us.

In this lesson we demonstrated simple Text to Speech (TTS) with this code.

Remember this program requires use of the AI Educational OS we flashed in LESSON #2.

from fusion_hat.tts import Piper
tts = Piper()
tts.set_model('en_US-amy-low')
msg = "Hi, I'm piper TTS. Are you Ready to Rumble? I am here to pump you up!"
tts.say(msg, stream=False)

from fusion_hat.tts import Piper

tts = Piper()

tts.set_model('en_US-amy-low')

msg = "Hi, I'm piper TTS. Are you Ready to Rumble? I am here to pump you up!"

tts.say(msg, stream=False)

As we say in the video, the first time you run the program you will get a permission error. This is because the model folders are inside a system folder and must be created as a ‘superuser’ using ‘sudo’. As shown in the video, you need to open a terminal window, and type in these commands at the command prompt (Put them in one at a time):

sudo mkdir -p /opt/piper_models
sudo chown -R pi:pi /opt/piper_models
sudo chmod -R 755 /opt/piper_models

sudo mkdir -p /opt/piper_models

sudo chown -R pi:pi /opt/piper_models

sudo chmod -R 755 /opt/piper_models

You only need to do that one time. Next time you run the program, all will work properly.

Then, in order to hear all the different voice models Piper offers, you can run this program, and each voice will introduce itself to you.

from fusion_hat.tts import Piper
from time import sleep  # Optional: for pauses between voices
device.enable_speaker()
tts = Piper()
# Get the dictionary of available en_US models
voiceDict = tts.available_models('en_US')
print(voiceDict)
# Flatten into a single list of full model names
allVoices = []
for voice in voiceDict.values():
    for var in voice:
        allVoices.append(var)
print(allVoices) 

msg = "Hello World. This is a test message."
for voice in allVoices:
    tts.set_model(voice)  # Auto-downloads on first use
    # Create a readable voice name (remove 'en_US-' and replace '_' with space)
    speakableName = voice.replace('en_US-', '').replace('_', ' ')
    print("Switching to: " + speakableName)
    print(voice)
    sleep(5)
    # Introduction in that voice
    intro = "This is the voice " + speakableName + "."
    tts.say("Hello There" + intro)
    #tts.say(msg, stream=False)
    sleep(.1)
print("All voices demonstrated!")
for voice in allVoices:
    print(voice)

from fusion_hat.tts import Piper

from time import sleep # Optional: for pauses between voices

device.enable_speaker()

tts = Piper()

# Get the dictionary of available en_US models

voiceDict = tts.available_models('en_US')

print(voiceDict)

# Flatten into a single list of full model names

allVoices = []

for voice in voiceDict.values():

for var in voice:

allVoices.append(var)

print(allVoices)

msg = "Hello World. This is a test message."

for voice in allVoices:

tts.set_model(voice) # Auto-downloads on first use

# Create a readable voice name (remove 'en_US-' and replace '_' with space)

speakableName = voice.replace('en_US-', '').replace('_', ' ')

print("Switching to: " + speakableName)

print(voice)

sleep(5)

# Introduction in that voice

intro = "This is the voice " + speakableName + "."

tts.say("Hello There" + intro)

#tts.say(msg, stream=False)

sleep(.1)

print("All voices demonstrated!")

for voice in allVoices:

print(voice)

Remember in these early lessons we are using this circuit to demo our programs. Please leave this circuit put together.

AI On the Edge, Tutorial

AI on the Edge LESSON 7: Homework Solution for Dimmable LED

April 18, 2026 admin

In this video lesson we present the solution to the homework assignment given in LESSON #6. Your assignment was to create a dimmable LED where the brightness of the RGB LED is controlled by the potentiometer. We are still using the schematic from our earlier project.

In this lesson, this is the code which we came up with:

#!/usr/bin/env python3
from fusion_hat.adc import ADC
from fusion_hat.pwm import PWM
from time import sleep
import math
potPin=0
myPot = ADC(0)
redPin=5
redLED=PWM(redPin)

while True:
    potVal = myPot.read()                     # 0 to 4095
    writeVal=100**(potVal/4095)-.005
    redLED.pulse_width_percent(int(writeVal))
    print(potVal,writeVal)
    sleep(0.1)

#!/usr/bin/env python3

from fusion_hat.adc import ADC

from fusion_hat.pwm import PWM

from time import sleep

import math

potPin=0

myPot = ADC(0)

redPin=5

redLED=PWM(redPin)

while True:

potVal = myPot.read() # 0 to 4095

writeVal=100**(potVal/4095)-.005

redLED.pulse_width_percent(int(writeVal))

print(potVal,writeVal)

sleep(0.1)

AI On the Edge

AI on the Edge LESSON 6: Digitial Out, Servos, Analog In and PWM on the Fusion HAT+

April 15, 2026 admin

In this video lesson we will become familiar with the FUSION AI HAT+, and we will learn how to do digital inputs and outputs, analog in readings, analog outputs, PWM and how to control servos. By the end of the lesson you should have the essentials you need to work with the Fusion HAT+. The schematic for the circuit we will be using in today’s lesson if below. We describe it in more detail in the video. The schematic is:

In the video lesson we demonstrated the following programs:

Digital output to blink an LED:

from fusion_hat.pin import Pin, Mode
import time
LEDPin=17
redLED = Pin(LEDPin,mode=Mode.OUT)
while True:
    
            redLED.high()
            time.sleep(.1)
            redLED.low()
            time.sleep(.1)

from fusion_hat.pin import Pin, Mode

import time

LEDPin=17

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

while True:

redLED.high()

time.sleep(.1)

redLED.low()

time.sleep(.1)

PWM Example to Control RGB LED Color and Brightness.

from fusion_hat.pwm import PWM
from time import sleep

redPin = 5
greenPin = 6
bluePin = 7

redLED = PWM(redPin)          
greenLED=PWM(greenPin)
blueLED=PWM(bluePin)
redLED.freq(200)
greenLED.freq(200)
blueLED.freq(200)

while True:
    for bright in range(0, 101, 1):
        redLED.pulse_width_percent(bright)
        sleep(0.01)
        print(bright)
    for bright in range(99, 0, -1):
        redLED.pulse_width_percent(bright)
        sleep(0.01)
        print(bright)

from fusion_hat.pwm import PWM

from time import sleep

redPin = 5

greenPin = 6

bluePin = 7

redLED = PWM(redPin)

greenLED=PWM(greenPin)

blueLED=PWM(bluePin)

redLED.freq(200)

greenLED.freq(200)

blueLED.freq(200)

while True:

for bright in range(0, 101, 1):

redLED.pulse_width_percent(bright)

sleep(0.01)

print(bright)

for bright in range(99, 0, -1):

redLED.pulse_width_percent(bright)

sleep(0.01)

print(bright)

Reading Analog Voltages on the Pi 5 Using the Fusion HAT+

from fusion_hat.adc import ADC
from time import sleep

potPin=0
myPot = ADC(potPin)       # A0–A3 on HAT

while True:
    analogIn = myPot.read()
    voltage=analogIn*(3.3/4095)
    print("Analog In: ",analogIn,"Voltage: ",round(voltage,2))
    sleep(0.1)

from fusion_hat.adc import ADC

from time import sleep

potPin=0

myPot = ADC(potPin) # A0–A3 on HAT

while True:

analogIn = myPot.read()

voltage=analogIn*(3.3/4095)

print("Analog In: ",analogIn,"Voltage: ",round(voltage,2))

sleep(0.1)

Controlling Servos With the Fusion AI HAT+

from fusion_hat.servo import Servo   # Import the Servo class for controlling servos
import time
# Import sleep for timing delays
panPin=2
tiltPin=3
servoPan = Servo(panPin)
servoTilt=Servo(tiltPin)
servoPan.angle(0)
servoTilt.angle(0)

from fusion_hat.servo import Servo # Import the Servo class for controlling servos

import time

# Import sleep for timing delays

panPin=2

tiltPin=3

servoPan = Servo(panPin)

servoTilt=Servo(tiltPin)

servoPan.angle(0)

servoTilt.angle(0)

Technology Tutorials

All posts by admin

AI on the Edge LESSON 10: Make Your Raspberry Pi Listen to You with Voice Commands

AI on the Edge LESSON 9: Make Your Raspberry Pi Talk with TTS

AI on the Edge LESSON 8: Text to Speech (TTS) on the Raspberry Pi

AI on the Edge LESSON 7: Homework Solution for Dimmable LED

AI on the Edge LESSON 6: Digitial Out, Servos, Analog In and PWM on the Fusion HAT+

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