More law firms are using speech-recognition tools

ntil a few weeks ago, Ilene Fleischmann, a vice dean at the University at Buffalo Law School, hadn't thought much about speech-recognition software.

She was content to dictate her work or to type things herself on her computer keyboard. It was how she had always functioned. But after breaking her wrist, Fleischmann had to shift her work process and, ultimately, her thinking.

Without the use of her wrist and hand, Fleischmann had to find a new means of creating correspondence and documents. Speech-recognition software emerged as an appealing option, enabling her to speak into a microphone and have her spoken words appear onscreen as if they had been typed, for distribution or printing.

After just a few weeks, Fleischmann has decided she's so satisfied with speech-recognition that she can't go back to standard typing. "I'll never give it up," she says ---- even after her arm heals.

Fleischmann is among the growing numbers of speech-recognition champions in the legal community ---- lawyers, paralegals, secretaries and administrative staff who have found that speaking their work (as opposed to typing or dictating it) makes them much more productive. They are part of a recent surge of interest in speech-recognition software. Until a few years ago, speech-recognition had more detractors than supporters. The software itself was problematic: It featured a limited vocabulary, accuracy was poor, and a time-consuming training process was required to generate the most basic sentences. But over the past few years, software developers have worked diligently to improve functionality, creating products that today only faintly resemble the first generation.

Today's speech recognition software typically has more than 300,000 words, terms and acronyms built in, with the ability to add industry-specific jargon as well. The latest versions of the software yield a 95 percent to 97 percent accuracy rate, mean­ing the words you speak will almost always be the words that appear onscreen and in your document. Using a statistical model to analyze sound patterns, the software interprets those sounds as commands or dictation based on three factors: acoustic data, or sound patterns; vocabulary; and a language model. Software manufacturers have been tackling improvements in all three areas in order to achieve better results for professional users such as lawyers.

As client expectations have risen with the quickening pace of business and infor­mation-sharing, so has the pressure to per­form. In a struggle to keep up with larger workloads and faster turnaround times, attorneys have invested in technology and productivity-improvement tools.

However, a slowing economy has forced some firms to retract, to reduce staff, and to shrink budgets. To do more with less, as the saying goes.

The convergence of those three trends ---- a faster business pace, increased reliance on technology and a weakening economy ---- is pushing speech-recognition software to the fore. The number of law firms turning to speech-recognition has increased exponentially in the last few months, a sign that its time has come. This technology is at the tipping point, the point just before widespread use and acceptance.

The slowing economy has put financial pressure on firms that may not have felt it before, causing managing partners to carefully evaluate each staff member's strategic contribution to the firm. In a down economy, one of the ways to gain a competitive advantage is to make smart investments in tools and capabilities that improve the firm's profitability ---- tools like marketing, specialist staffers and productivity improvement devices.

Many firms have discovered that speech-recognition software eliminates unneces­sary expenses like traditional transcription and the time administrative personnel devote to transcribing documents on-the-job. Speech recognition eliminates the physical typing step of traditional transcription or manual document creation, freeing up valuable support staff time, which can be repurposed. On average, most firms receive a return on investment in approxi­mately two months.

The real benefits of speaking instead of typing, however, come with the ability to multitask hands-free, as Ilene Fleischmann discovered unexpectedly.