This is the how. The why is in the first post, so here's the short version. I pitch in a men's league, I wanted to call my own pitches, and I built a DIY pitch caller like PitchCom, except mine only goes one way. It clips to my belt. I pick the pitch, press a button, and my catcher hears my own voice say "fastball" in the earbud.
A lot of people asked how to make one. This is how it works, including what went wrong. The code and the full build guide are on GitHub (link at the bottom).
If you're wondering why this took a while, a patent question came up after the video. I took the time to read through the patents and talk with a patent attorney before sharing anything. More on that below.
Never coded? I wrote a step-by-step guide for coaches who have never touched code or GitHub.
How It Works
The whole thing is three parts:
- An M5StickC Plus2 on my belt. It's a little stick with a screen, two buttons, Bluetooth, and a battery. About $20 to $30, depending on where you find one.
- A cheap Bluetooth earbud in the catcher's ear, like this one.
- A visor clip off my garage door opener to hold it on my belt. (Free, and the garage still opens.)
I recorded myself saying fastball, curveball, slider, and changeup on my computer, and those four clips live on the stick. When I hit send, the stick plays that clip over regular Bluetooth audio. Think of it like your phone playing music to your earbuds. The earbud has no idea it's at a baseball game. It plays whatever shows up.
I didn't invent any hardware. I found a cheap device that sends Bluetooth audio, programmed it, and recorded my voice. That's the whole trick.
Why There's No Location
No need to call location if you don't know where it's gonna go.
But what you record and send is dealer's choice. It doesn't have to be the pitch name. Use a number, a code word only your catcher knows, or put location right in the call, like "fastball in." Each one is just another clip.
Check Your Bluetooth First
My first try didn't work. I asked Claude to plan it and followed along (that story is in the first post). The reason it failed is worth knowing before you buy anything.
Bluetooth comes in two flavors. Bluetooth Low Energy is like texting: short little messages that sip battery. Bluetooth Classic is like a phone call: a steady stream, which is what audio needs. Earbuds take the phone call. The chip I started with could only text, so the earbud never connected.
If you build your own, check for Bluetooth Classic before you buy anything.
The Screen
The stick sits on my belt with the screen facing out, so once the call goes out, the screen only says "PITCH SENT." No pitch name, no history. The next press always starts back at fastball, so I can run it by feel and it looks the same every time.
How It Tested
It still hasn't been used in a real game, and it's hit or miss at the full 60 feet 6 inches (the testing story is in the first post).
Some of this is on the hardware. After the first build went wrong, I went cheap, because I didn't have money to waste on an experiment. A cheap stick and a cheap earbud have their limits.
Known problems:
- Range. It gets shaky at full distance. The stick never turns up its Bluetooth power, and my body sits between it and the catcher.
- Volume. My recordings were quiet. Record yours louder and closer to the mic.
- Battery. The screen and the audio stream run nonstop, so it wants a charge between innings.
- Other devices. If the catcher's phone has Bluetooth on, the earbud might grab the phone instead.
That's why I'm not selling it. It wouldn't be right to charge people for something that's hit or miss at full distance.
The Patent Guy
After I posted the video, one of PitchCom's co-owners replied. He's also a patent attorney. His opener: "Have you considered the patent rights of PitchCom and whether you would be violating those rights and setting yourself up for expensive litigation?"
Fair question. So I asked him which rights. By then I'd already started going through PitchCom's patents, and nothing I'd read matched. His answer: "Review our 10 issued US patents covering our system." So I finished all 10. Still nothing.
I asked again how I violated anything. "I'm not going to analyze your designs and do your work for you."
Then this: "Our attorneys have not performed an infringement analysis on your design since they don't have it." In the same reply he said he "tried to help you before legal action was necessary." Then he was done: "I'm not going to discuss this more in an open forum."
So he never named a single claim. His own attorneys hadn't looked at it. And legal action still came up.
Here's the difference. In every one of their patents, the earpiece or receiver stores the calls, and the sender transmits a short code that tells it which one to play. OpenSigns works the other way around. My calls live on the belt device and stream as plain audio to an ordinary earbud that stores nothing. Their earpiece is the library. Mine is just a speaker.
OpenSigns isn't affiliated with PitchCom.
One more thing before you build. That's my read of their patents, not legal advice. If you build this, change it, or build something similar, I advise you to check the patents first and make sure you don't infringe. Don't build, use, or sell anything that infringes their patents. The 10 patent numbers are in the README.
Build Your Own
Check your league's rules before you use it in a game.
The code, my four recordings, a tool to turn your own recordings into something the stick can play, and a step-by-step build guide are all on GitHub. The code is GPLv3 and my recordings are CC BY 4.0. Record your own voice, change the earbud name in the code, and you're in business.
Building this has me looking at other cheap devices and wondering what else they could do on a ball field.
If you've never coded, start with the coach guide. It walks you through the whole build with an AI chat as your helper.