Look, Your Honor -- No Hands!
Melissa Macaluso transcribes 200 words per minute without ever lifting a finger.
She is a court reporter, but instead of typing the proceedings, she repeats the testimony word for word into what looks like a gray oxygen mask with foam rubber around the mouthpiece. The device, called a stenomask, is hooked up to a laptop with speech-recognition software, and her voice, muffled by the mask, is only a faint hum. Her words are recorded by a microphone embedded in the mask and instantaneously translated into text on her computer screen.
"You hear it and you spit it out," she said. "You don't listen for context. You don't listen for detail."
The ease of stenomask reporting, or “voice writing,” as practitioners call it, is leading to an upheaval in the world of transcription. Would-be transcriptionists are shunning traditional machine shorthand--using those narrow keyboards that court reporters peck away on--and are turning instead to easier-to-learn voice writing. The number of transcriptionists enrolled in traditional court-stenography courses has dropped by half since 1992, and training schools across the U.S. are closing as voice writing takes over.
“There are a lot of different forces at play,” said Peter Jepsen, former director of captioning services for the U.S. Senate, where he provided closed captions for televised broadcasts of Senate proceedings using machine shorthand. In 2000, he saw a demonstration of speech-recognition technology and immediately made the switch, and now he teaches voice writing. “If anything, voice writing will be the salvation in the profession," he said.
Macaluso learned stenomask reporting at Audio-Rite Corporation, the only place in New York City that teaches voice writing. Claire Block, the company's owner, started out 25 years ago with Lex Reporting Inc., a provider of independently contracted court reporters, when machine shorthand (or stenography) was the industry standard for court reporting and television captioning.
In recent years, however, Block had trouble attracting stenographers, and court-reporting schools were closing. In 2004 she bought speech-recognition software called Audioscribe and began training people in voice writing. “Anybody can learn it,” Block said. “You only have to know how to speak.”
Since then, Audio-Rite has trained nearly two dozen people in voice writing. Advances in speech-recognition software have made the field a more attractive career path, she said. “This is the next generation of court reporting. I really believe that.”
Jepsen, the former U.S. Senate captioner, now teaches writing at AIB College of Business in Des Moines, Iowa, which offers one of a growing number of academic programs to train students in voice writing. “Stenotyping is hard to learn,” he said. “We had people who were here three, four, five years, and they weren’t progressing.”
Last year, Jepsen talked AIB into trying a pilot program in voice writing for broadcast captioning, recruiting students from the stenography program. He said they were able to pick up voice writing faster and more easily than machine shorthand. The school now offers the program to the general public, and 27 students are now working to earn a voice-writing degree. “People coming out of high school now have different goals and dreams,” Jepsen said. “They don’t want to learn a stenotype machine.”
In fact, the number of stenography students has dropped dramatically, from a peak of 14,000 in 1992 to 7,000 in November 2006, according to the National Court Reporters Association in Vienna, Va. Only 10 percent of those enrolled are expected to graduate. The number of schools teaching stenography has declined too. In 1992 the association recognized 99 stenography schools in the U.S.; that number is now down to 74.
In New York state, only five new licenses have been granted to shorthand reporters so far this year, from 26 in 2002, according to the New York State Education Department's Office of the Professions, which grants the licenses.
The length of time required to complete a stenography program also factors in the shortage of court reporters. It takes an average of 33 months to learn machine shorthand, versus six to nine months for voice writing. At the New York Career Institute, the only school left in Manhattan that teaches machine shorthand, three years of training will cost more than $35,000. By contrast, Block’s Audio-Rite program costs $20,500, but because it's not a school, there is no financial aid.
While voice writing has recently become popular, it was actually invented in the 1940s. Horace Webb was a shorthand court reporter frustrated by lawyers and witnesses who talked too fast. He decided to find a way to repeat everything he heard into a microphone, record it and transcribe it all later.
He experimented with a microphone and a cigar box, tomato cans, bottles, and boxes of all shapes and sizes, but the recordings were unintelligible. Finally, an acoustic engineer diagnosed the problem as “reverberating sound waves,” which made his voice bounce around inside the box and echo into the microphone.
Webb’s low-tech solution was to attach a rubber face piece from the Air Force to a hollowed-out coffeepot stuffed with rags. That dampened the sound waves, and his voice was audible in the recording yet silent enough not to interrupt court proceedings. Stenomask reporting was born.
Webb died long before voice writing started to gain traction. When Linda Drake first encountered voice-recognition software in 1997, few people had computers fast enough to make it work. Now “the chip technology is better,” said Drake, a board member of the National Verbatim Reporters Association, the official organization of voice writers, which has 1,000 members. “The software is now at a point where it can figure out context, not just the sound of the word,” she said.
Voice writing is becoming widely used in captioning and is now considered an acceptable form of court reporting in 26 states and the District of Columbia. It is not permitted in the remaining states because of certain administrative codes that legally define court reporting as written or typed shorthand.
Arizona and Nevada became the latest states to allow voice writing after a lawyer turned voice writer petitioned the court-reporting boards of both states. The National Verbatim Reporters Association’s Drake helped get the court-reporting standards in those states changed through demonstrations and education, convincing the court-reporting boards that voice writing was up to the task. Nevada approved voice writing in 2005, while Arizona accepted it this year.
“There was resistance from stenographers,” Drake said. “I think it was the fear of new technology.”
Indeed, there is resistance to voice writing from the National Court Reporters Association. “There are still some issues with voice writing,” said Marshall Jorpeland, director of communication for the group. “Voice writing is good up to 200 words per minute, but it doesn’t go smoothly above that.” Certified court stenographers, he said, type a minimum of 225 words per minute.
Membership in the association has steadily declined over the last 10 years, from 31,000 to 23,000 in 2006, but it's still far more than the membership in the voice writers association.
Jorpeland doesn’t rule out the possibility of allowing voice writers to join his organization, but that would require a change in bylaws. “Certainly we have thoughts about where voice writing is,” he said. “Some people think it’s the same job. But some people think stenograph reporting is what we represent, and that’s what we should focus on.”