Tips & Tricks interact with DIY voice assistant
We have a cool STEAM educational toy called Circuitmess Spencer, who is also a personal voice assistant.
Discover the wonders of voice recognition technology – a revolutionary way to interact with your devices using just your voice. Experience the convenience of hands-free control, improved accessibility, and personalized assistance in various industries.
Immerse yourself in the world of voice recognition technology and unlock the effortless ways to utilize your voice for seamless interaction with everyday technology. Discover a fun, interactive, and hands-on approach to master the art of voice-controlled technology in your daily life.
What you can do with CircuitMess Spencer
CircuitMess Spencer is a DIY voice assistant that you can build and code yourself. It is a fun and educational project that will teach you about AI, voice recognition, IoT, and speech synthesis. Spencer also has a funny personality, tells jokes, and finds new ones online.
Build your own voice assistant :
- The first thing you need to do is to assemble Spencer using the components and tools provided in the box.
- You will learn how to solder and connect different parts, such as the LED grid display, the speaker, the microphone, and the big red button.
- You will also learn about the brain board, which is the microcomputer that powers Spencer and runs your code.
- The build guide will walk you through every step with detailed instructions and photos included in the box.
Code new functions
Once you have built Spencer, you can start coding new functions using CircuitBlocks or C/C++. CircuitBlocks is a graphical programming interface that lets you drag and drop blocks of code to create your program. C/C++ is a text-based programming language that gives you more control and flexibility over your code. You can use either one or both to program Spencer according to your preferences.
Some of the things you can code Spencer to do are:
- Ask about the weather forecast for your area
- Hear a joke
- Ask him to sing you a song
- Set a stopwatch
- Make Spencer display custom animations
- Laugh at his corny popular culture references
- Talk to Spencer using voice commands
Use CircuitBlocks and C/C++ to create your own functions and trigger them with the big red button or your voice.
Learn about AI and IoT
Spencer is not just a toy, but also a learning platform that will introduce you to the concepts of artificial intelligence and internet of things.
- You will learn how Spencer uses voice recognition and speech synthesis to understand what you say and respond accordingly.
- You will also learn how Spencer connects to the internet and uses online services such as OpenWeatherMap and JokeAPI to get information and data.
- You can also explore other online APIs and services that Spencer can use, such as Wolfram Alpha, Wikipedia, Spotify, Twitter, etc.
- You can make Spencer answer your questions, play music, tweet something, or anything else you can think of.
This is how you can Make Spencer listen and repeat.
With CircuitMess Spencer, the possibilities for voice interaction are endless. Harness the full potential of this DIY voice assistant as you unlock the captivating feature of making Spencer listen and repeat. Imagine the excitement of having your own voice echoed back to you, creating a world of interactive fun and learning. Let’s dive into how you can bring this enchanting experience to life!
Activate the Voice Recognition Mode:
Ensure that CircuitMess Spencer is assembled and powered on, ready to embark on this thrilling journey.
Explore the voice recognition mode, which allows Spencer to listen and respond to your commands.
Speak Clearly and Command Spencer:
Once in voice recognition mode, speak clearly and distinctly to provide instructions or prompts to Spencer.
Give commands like “Spencer, repeat after me” or “Hey Spencer, echo my voice” to initiate the listening and repetition process.
Enjoy the Delightful Interaction:
As you speak, Spencer’s advanced technology will capture your voice and repeat it back to you, creating an engaging and interactive experience.
Listen to your own voice being echoed by Spencer, and immerse yourself in the joy of playful interaction.
Explore the Learning Opportunities:
Making Spencer listen and repeat is not only entertaining but also an excellent learning tool.
Enhance your communication skills, pronunciation, and verbal expression by engaging in conversations with your voice assistant.
Unleash your imagination and make CircuitMess Spencer an active participant in your daily activities. Whether it’s practicing languages, storytelling, or simply enjoying the entertainment value, making Spencer listen and repeat adds a whole new dimension to your relationship with this incredible DIY voice assistant.
For the ultimate coding challenge, let’s make Spencer listen to what we say and repeat that same sentence with his funny robotic voice.
We can just continue with the code you have made in the previous chapter for this one.
Firstly, delete the “synthesize and say” block by right-clicking on it and pressing the delete option.
So now we need the “start listening” block.
Place it here:
With this code, every time you press its red button, Spencer is going to check if WiFi is connected, play the loading animation, and start listening to what you say.
When Spencer is done listening, let’s play an animation so that we know it’s now processing our voice command.
Now find the “when speech gets processed” event block. This block will be triggered once Spencer processes your voice command and receives a transcript from the server.
Drag and drop it onto the drawing area:
Since we want Spencer to repeat what you’ve said to him, let’s find the “Synthesize and say” block:
Place it here since we want to make Spencer repeat your words as soon as your voice gets processed:
We want Spencer to say the exact thing that you have said. That’s why you need to take the red “intentResult->transcript” value block and drop it into the
“synthesize and say” block.
Finally, let’s play a talking animation when Spencer’s voice sample gets generated and make him wink once he’s done talking.
Awesome, your code is done now.
Save it and run it on your Spencer.
Smack Spencer’s head and tell him something (i.e., “Hello Spencer, you are my friend”). Spencer will process your voice command and repeat it with his robotic voice.
Some historical information about early voice recognition technologies:
- Audrey: In the 1950s, Bell Laboratories created a device called Audrey, which could recognize spoken numbers. It was limited in functionality and could only understand digits.
- Harpy: In the 1970s, the U.S. Department of Defense’s Advanced Research Projects Agency (ARPA) developed Harpy, a speech recognition system designed to understand and respond to voice commands. It was capable of understanding approximately 1,000 words.
- IBM Shoebox: In 1962, IBM introduced the “Shoebox” machine, also known as IBM 7040. It was an early speech recognition system that could recognize 16 spoken words and digits.
- Dragon Systems: In the 1980s, Dragon Systems (now Nuance Communications) released a speech recognition software called DragonDictate, which allowed users to dictate text into a computer.
While these examples highlight the development of voice recognition technology, they were not voice assistants in the sense of interactive digital assistants like those we have today. The concept of voice assistants as we know them today started to emerge in the early 2000s with the introduction of products like Siri (2011), Google Now (2012), and Amazon Echo with Alexa (2014).
First sight for voice commands
Radio Rex was indeed a popular toy dog introduced in the 1920s, particularly in the United States. It was a battery-powered mechanical toy that resembled a dog and had some interactive features. When a person spoke the name “Rex” or other specific commands, the toy would respond by wagging its tail, moving its head, or performing other simple actions. It was primarily a mechanical toy and did not incorporate voice recognition or advanced interactive capabilities like modern voice assistants.