Baseball-Reference’s Flawed Audio Search
If you are a baseball fan, you’ve got to love stats, and that means you’ve got to love Baseball-Reference.com. I’ve wasted many, many, many hours on this site because of course I must know how many home runs Bruce Boisclair hit in his Mets career (it’s ten, by the way).
During one of my many daily visits to the site recently, I noticed a little microphone in the search box. I guess they added a voice search function, so I decided to test it out. I said loudly and clearly “Tom Seaver,” and the words “I Am Super” popped up. So I said it louder and more clearly, and his name came up properly.
I decided to try a more difficult name. To my surprise, “Mike Piazza” came up just fine. For the ultimate test, I tried “Carl Yastrzemski.” I did not get “Carl Yastrzemski” — I repeatedly got “Carl Your Strengths Key.”
So proving I have entirely too much time on my hands, I spent the next half-hour testing the system with names that are not easy and I found that the folks over at Baseball-Reference have a little more tweaking to do on their audio system. Here is what I got (and these are all true):
Lee Mazzilli:
Leave Early
Sibley Missouri
What Youdoing
Linux
Emily
Rusty Staub:
Rusty Stop
Ron Guidry:
Wanted Delivery
Ron D 3
Ron is Reese
Ron January
Ron You Treat
Ron Geary
Jon Niese:
John Knits
Chris Capuano:
Chris Kathy 10
Chris Happy Wanderers
Chris Can You Wanna Know
Tim Teufel:
Tim Couple
Jose Reyes:
Raiders
So is a Radius
Old is a Race
Who is a Race
Joe DiMaggio:
Jovan as You
Moises Alou:
What is a loser
What is a Loop
Mississippi Allure
Voice is a Loom
Ralph Kiner:
Ralphs Kinder
Ralph Carter
The Kiner stuff is fitting because he never got anyone’s name right (remember “Gary Cooper” for “Gary Carter” and “Darryl Throneberry?!).
Oddly, the most difficult name in the majors today, Jarrod Saltalamacchia, came up fine — a bit misspelled, but I was expecting something like “Jar Head Salt Allah Smack You.”
So if you have some time to kill, go on the site and have some fun before they fix it. And by the way, “Bruce Boisclair” came up “Bruce Eauclaire.”
The audio search is all on google. They take the audio run it through their servers and then return it to us. I’m sure there are lots of names it struggles with.