There are two prevailing schools of thought when it comes to the gold standard of machine shorthand stenography in United States legal proceedings. There are those that truly believe in the standard. There are those that give it lip service, only ever talking about stenography when pressed or pressured. Of course, there’s a third school of thought in the people that can’t or won’t spend much time thinking about why we still use our chorded stenotype keyboard design over a century after its development. For the third schoolers, we use QWERTY layouts despite that design being over a hundred years old too. It’s easy to imagine why: 1. There’s a market for it. 2. No technology has come along that is more intuitive and better.
I recently had an experience where I had to pick something off of an audio recording painstakingly in transcription mode. It gave me a lot of insight into where stenography’s superiority comes from. It’s in the room control. Some people are always going to be able to speak faster than we can “write” or type. You throw a stenographer into a situation where they have no room control and the participants are speaking above the stenographer’s skill level, and what do you get? You basically get digital court reporting / recording. The stenographic notes are a useless game of fill in the blank.
For the last twelve years that I’ve been in the industry, companies have been pushing reporters to interrupt less. I get it. Just like anybody else, lawyers don’t like to be interrupted. The loudest complaints were probably from the ones that are most self-important. The companies likely sought to end complaints by telling stenographers to let the audio catch it. But every time we do that, we risk record degradation “Didn’t understand that when they said it, don’t understand it no matter how many times I replay it.” It also increases the amount of time we have to spend on the matter due to re-listening to testimony rather than having it clearly in our notes. Since many depositions go unread until there’s a motion to be filed or trial’s coming up, the number of complaints related to poor transcript quality will likely always be lower than the number of “your reporter interrupted me” complaints. This skews the world the non-reporter owned agency lives in. Make the customer happy and things will work out. Just hope they don’t need whatever was inaudible or unintelligible to make their case.
That’s a major problem for digital, and I am not the first one to write something like this thanks to Jean Whalen. You have audio monitors that may or may not know anything about legal transcription listening for issues that they anticipate the transcriber will have. By removing the ability of the person responsible for the transcript to interrupt, you increase the chance of serious errors. Throw away all my prior calculations. The answer is really that simple.
From a productivity standpoint, room control makes a big difference. I’ve timed myself no audio versus heavy audio use, and I personally can be an astounding 12x slower putting together a transcript when heavy audio use is involved. This is why collectives like Ana Fatima Costa’s Speak Up For The Record group are so vital. In some jurisdictions, there is no mandatory license. There is no legal standard. Our newbies and veterans alike are connected to best practices through the stories and experiences we share amongst ourselves and the encouragement we give each other to be better. Let that be my share: We will not be attracting anyone to this field if they’re peeling things off audio in the name of “our client doesn’t like to be interrupted.”
The Lip Service School
More mainstream legal news has been picking up on the fact that there’s an ongoing debate. I’d like to share some highlights from the article “Glitches Still Persist In Digital Court Reporting Tech” by Steven Lerner, Law360 Pulse.
- “…90 hours of testimony digitally recorded in a trial in the Northern Mariana Islands in 2008 resulted in poor audio quality and transcripts that were deemed unreliable and inaccurate.” It’s worth mentioning, but since it was so many years ago, it’s a minor point.
- Planet Depos told Law360 Pulse that the problem with a 285-page transcript in Maryland was not the technology, but rather the setting of a public hearing where they were unable to control audio quality, overlapping speakers, and random unidentified speakers scattered across a large room. This goes directly to my points about room control. If we are not serious about speaking up when the record is in danger, we are not serious about record accuracy. Customer education is going to be this decade’s biggest challenge.
- Brian Jasper, an attorney at Thomas Law Offices PLLC told Law360 Pulse “the technology was a problem, and it interrupted the deposition. I don’t scrutinize the depositions for perfection, but as an attorney, I have much more confidence in a stenographer because they are taking it down in real time.” This speaks to my point on room control. We generally know when we’re not getting it.
- The article talks about the Stanford study where voice recognition by Apple, Google, Microsoft, IBM, and Amazon was tested. Error rates for black men were over 40%. I’m happy that this is getting more attention, because the adoption of automatic speech recognition (ASR) into legal transcription can really hurt equality and quality in general.
- Stenograph, through Anir Dutta, claimed the average wait times for customers is seven minutes. This conflicts with reports at the end of last year that wait times for some were over a half an hour. Anir Dutta is quoted as saying “if that means that that customer is going to go on Facebook and make it so that everybody thinks that our average hold times are tremendously high, I think it’s unfair and frankly malicious.”
- Lisa Migliore Black is quoted. “After 25 or more years of always keeping my Stenograph support contract up to date so that I would have the most current software advances, I let my support contract expire in January of 2022 due to long hold times with technical support and their failure to resolve the problems I was experiencing over the course of several months.” “My perception as a customer is that Stenograph is pulling too many available resources to develop the ASR side of their business.” I have to say I’m with Lisa. after over a decade of using CaseCAT, I’m very slowly teaching myself Eclipse, because being married to Stenograph just comes off as risky to me. The company seems obsessed with being at the helm of an evolution in court reporting that may never actually happen.
- Dutta stated 80% of the company’s investment is still in stenography and that it is a “false narrative” that going into digital court reporting is shifting its focus. He’s quoted saying “If Apple started making iPhones, does it mean that they make substandard laptops?” Again, this goes against what has been documented prior, a drop in customer service.
- Asked about the Stanford study, Dutta stated “People can quote studies from three years ago….” “…technology moves a million miles every three months.” This is demonstrably false. There’s a patent from 2000 showing 90% automatic speech recognition (ASR) accuracy was thought to be possible. The 2020 Stanford study showed accuracy lower than 80%. Is there anyone on Earth that believes 90% to 80% over the course of two decades is technology moving a million miles every thee months? ASR has improved. But it largely depends on who’s speaking and how good the audio is. I also find it humorous that Dutta takes exception to a 2020 study being cited when the entire basis for digital court reporting infiltration is Jim Cudahy, Speech-to-Text Institute, and a 2013-2014 Court Reporting Industry Outlook. Odd that an entire industry should shift focus for something that was done almost a decade ago and never adjusted for but should pay no mind to current events because “tEcHnOloGy.”
It’s a very interesting time to be in court reporting because nobody knows what happens next. Do the shot callers realize they’re wasting a lot of money trying to create a market for digital court reporting and start investing in the training of stenographers that will make them consistent profits? Will there be a breakthrough technology that renders stenography obsolete? Will our shortage get worse? Will our adoption of remote technologies compensate for the uneven distribution of court reporters across the country?
The data we’ve got doesn’t point to replacement. Until there’s a magic box that does everything, humans will be required to control the room, and it never gets more efficient than someone turning the speech into text right then and there with 95% or more accuracy. I’ll speculate that technology like CoverCrow will become more polished, mainstream, and accepted in helping with stenographer shortage woes. Agencies say they’re having coverage issues, and from what I understand, CoverCrow aims to work collaboratively with companies rather than cutting them from the equation.
As it stands, stenographers have a huge say in what happens next. Why?
- There’s a market for it. 2. No technology has come along that’s more intuitive and better.