You’re watching a ball game or a tv show and text is scrolling along at the bottom of the screen, just a few seconds behind the actual action.  There are some misspellings and occasionally a screwed-up word.  If you watch, you’ll see that the words identify the speaker in the conversation.  In a crowded, noisy restaurant; at home with the sound turned down; or if you are hearing-impaired, you can follow the action and discussion on the television.  It’s a little bit like those old Greek or Japanese fantasy/science fiction shows, where everything is a bit out of synch, but still it catches every word of the dialogue.

How does this happen?  Is it new and wonderful voice-recognition technology?  If so, how does the computer tell who is speaking, as well as what they’re saying?

It’s not a computer.  The person who is tracking and writing the dialogue so it can scroll across the screen is a court reporter, also called a stenographer.  A court reporter with a computerized stenotype machine can write over 200 words per minute with accuracy of nearly 99%.

I’m married to one.  Sometimes it’s pretty scary.  You can NOT be stupid and be a court reporter.

For centuries (actually, millennia) transcriptions were done in shorthand.  Cicero’s slave, Marcus Tullius Tiro, invented shorthand transcription in Latin around 63 B.C.  An English shorthand was developed in the 11th century, as a secret language to keep records of various kinds of nefarious and probably dangerous proceedings.  Charles Dickens used shorthand as a law clerk, and then as a free-lance court reporter.  Stenography machines were first developed around 1890.  In the early 20th century court reporters started using recording devices so they could keep an audio record to refer to for accuracy.  IBM attempted to computerize stenographic language-translation machines for the Army in the 1950s.  Then, demand grew for computerized steno machines, leading to the system used today which you have seen in movies, on television, and in the courtroom.  The court reporter sits at a desk with something that looks like a typewriter and takes down the testimony as it is spoken … every word of  the witness, the lawyers, the judge, and in some cases, the jurors.


As you can see, the steno machine is not a full-size typewriter, and these days it is connected to a computer.  The keyboard looks like this:

stebi pad

And the machine looks like this:

steno machine

There’s a particular twist to the court reporters’ skill.  They listen to phonetics, not spelling.  They don’t even really listen for context at the time they are “writing” the transcript.  The steno machine has a small screen that scrolls the transcript in English.  By connecting to his computer, this reporter is also able to see a larger part of the transcript as he is writing the testimony.

With improvement in computers, internet, bandwidth, and communications, court reporters can now provide special services like “real time”, in which a lawyer can see the transcript as it is being written; and “real time remote”, in which a lawyer in a different office or different city can see both the witness and the transcript simultaneously.  Technology also allows a lawyer with an appropriate connection to ask questions in a deposition or courtroom proceeding from somewhere other than the courtroom or deposition site.

Why do we go to all this trouble?  It’s very simple; we use court reporters for all kinds of proceedings so that we can have an accurate record of what was said.  Jury trials, appeal hearings, depositions of witnesses, and administrative hearings all require accurate written transcripts that can be referred to later.

For example, in a civil case, if I take your deposition, I might use the transcript of your deposition testimony to cross-examine you with at trial.  It would go something like this:

“Mr. X., you are saying Y about Z, correct?”  A:  Yes.  “But Sir, would you please read to me from the transcript of your deposition, under oath, on April 18, 2016?”  A:  “At that time I said B about C.”

“Aaah, you said B about C then, but you’re saying Y about Z now, isn’t that correct?”  A:  “Yes, that’s right.”  Or, “I must have been confused” … “I don’t remember.”  “It’s been a long time.”  It gets really bad if you say something like, “I must have been on drugs when I said B about C.”

That’s called impeachment:  confronting a witness with an inconsistent prior statement.  Impeachment is especially effective when it’s recorded in a stenographic transcript, under oath, by a certified court reporter.  There’s probably nothing more exciting and fun for a lawyer than setting, laying, and springing that trap on a witness who’s trying to win his case by being careless with the truth.  Don’t be the witness who says Y about Z after previously testifying to B about C.

Another use for transcripts is appeals.  Appeals are decided based on the “Record on Appeal”, which includes exhibits and transcripts from the trial court.  The appealing party has to pay for the cost of transcription of all parts of the record designated by any party to the case.  If the record is not there for the court to review, the issue it refers to can not be appealed.  A friend of mine won a magnificent victory a year or so ago for some people whose farms and homes were destroyed by a wildfire started by an oilfield pipeline company.  The court reporter lost the entire transcript somehow.  That was a 15 million dollar verdict and weeks of trial work down the drain.  Sucks to be you when that happens.  Sucks to be the court reporter’s malpractice insurer too, I guess.

But fortunately, it’s very rare.  Court reporters are by nature extremely diligent, intelligent, and detail oriented.  I would say they’re OCD, obsessive-compulsive disorder, but like I said my wife is a reporter and she’s likely to read this blog.  So, they’re just OC, not OCD. Right, baby?

Getting back to the nuts and bolts of reporting, court reporters work at least a couple of hours for every hour they put in “writing” the live transcript.  After creating the original and saving it to their computer, they “scope” the job.  The reporter or a hired “scopist” goes through the transcript and puts things in context — lines, paragraphs, interruptions, spelling, and stuff like that.  They’re also checking to see that they used the correct variant of words that sound the same, but are spelled differently and may have substantially different meanings.  Words like where, were, and we’re each have their own individual set of keystrokes.  The modern steno machine has “artificial intelligence” software that recognizes phonetic sounds and is able to string them into words.  Remember, the software doesn’t “type” English; it converts keystrokes into words, and each word has its own unique set of keys.

Transcripts are not exactly cheap.  They charge by the page, double-spaced, 25 lines per page.  Rush jobs (produced in less than a week), extra copies, real-time, and remote real-time all add to the cost.  A reporter will usually write about 250 pages a day in trial, so a one week trial transcript is likely to cost over $5000.00.  This is great when you’re married to the reporter, but not so great when you’re paying the bill for the transcript.

There’s actually a shortage of certified court reporters these days.  It’s one of the secret professions that an intelligent person can break into and make pretty good money.  In Oklahoma, some of our district courts in counties with smaller populations cannot keep reporters on staff, and are trying to fill the gap by hiring reporters on a per diem basis.  Federal court reporters have a pretty nice gig, since they get a decent salary, the opportunity to make extra money on transcripts for appeals, and federal health and retirement benefits.

Court reporters are absolutely essential to the proper functioning of our legal system.  Computerized voice-recognition systems can’t fill the gap, because they don’t have the accuracy, or ability to distinguish different speakers, necessary to create a good transcript.  And besides, at least one unfairly-sentenced defendant had to be happy there was an accurate transcript:

COURT: All right.  Mr. Shaw, let’s have an understanding here.  Listen to me.

DEFENDANT: I brought –-

COURT: I’m asking you to listen to me, and I’m telling you what is going to happen.

DEFENDANT: You’ve already denied me due process of law, sir.

COURT: Get the duct tape out.

(Whereupon, court remained in adjournment.  At which time the Defendant was bound and gagged.)

This was recorded accurately in the transcript in Shaw v. State, 846 S.W.2d 482 (Tex. App.—Houston [14th Dist.] 1993), and therefore was available for review by Texas’s criminal appeals court.

The appellate court held that defendant had been unfairly prejudiced as a result of sitting in front of the jury for the duration of the trial while bound and gagged with duct tape: “Even after considering the appellant’s prior criminal record we find this sentence (99 years in prison) somewhat on the severe end of the spectrum for stealing cartons of cigarettes from a grocery store.”

With which, your honor, I concur.