By: Danielle Chazen

The opinions expressed in this column are those of the author and do not reflect the perspectives or positions of STTI. In order to facilitate discussion and debate on issues and developments facing the speech-to-text industry, STTI welcomes the contribution of thoughtful commentary and content from those with relevant expertise and perspectives.

Speech-to-text technology and artificial intelligence are two terms that can conjure both mystery and hesitation. Today, the business application of these technologies is often grounded in more day-to-day tasks to reduce costs and respond to the needs of more access and efficiency.

However, AI-based speech-to-text, or automatic speech recognition (ASR), technologies are accelerating in their applications within the legal industry.

AI and ASR to the Rescue
The Wall Street Journal, among others, has pointed out a very real problem about the attrition in available court reporting resources, especially stenographers. There were 18% fewer stenographers in 2018 than there were three years earlier. Fewer individuals are entering the field, yet the demand for their services continues. The problem will not get better with time, as longtime stenographers continue to retire.

Court reporting and legal transcription are therefore primed to adopt AI and automation. Courts themselves are seeking to reduce costs and make access to court services, such as transcription, more affordable for litigants. There’s also the lack of stenographic services for below-the-line cases (low-value litigation, short depositions) and robust e-discovery.

ASR for Legal Transcription
ASR engines are being used to expedite an array of legal industry tasks. Court reporters and agencies are employing speech to text software for accurate and effective legal transcription. This software utilizes artificial intelligence and machine learning to produce transcripts. The machines often self-learn to “get smarter” and recognize more terms to provide more accuracy with each use.

Some AI technology transcription providers fact check the work of their automatic speech recognition machines with human editors. Verbit utilizes two human editors per project to obtain 99% accuracy, produce transcripts that are top-notch, and ensure legal parties involved are provided with technology that promotes a fair legal system.

In a legal setting, customization is also key. Many providers, including Verbit, specialize in legal and understand the needs as they pertain to legal proceedings. Clients are provided with opportunities to fully customize the process with templates. Users are also given the ability to upload information before the machine is used so the machine immediately recognizes complex legal terminology or names of relevant speakers.

Benefits & Takeaways
For court reporting agencies, the application of AI-based ASR machines to transcription efforts is paramount. It can result in reduced overhead costs, a turnaround of transcripts faster, and increased operational capacity to serve more clients. Many professionals are benefiting from the ability to separate the audio capture of legal proceedings from the production of the transcripts. This unbundling of the workflow and new reliance on AI technology speeds up the process.

Benefits to litigators also include access to searchable audio, on-demand transcripts, and potentially lower costs of service. Court systems also may benefit from digital recording and transcription for reduced operating costs, as well as faster turnaround times for litigants.

However, technology will not replace the human factor. Combining technology with human-delivered services offers a way to address scalability challenges in the face of a dwindling supply of stenographers while still ensuring quality. For the highest accuracy, an individual still needs to listen to the technology-captured transcript for the record.

Here is some more helpful information on Verbit’s AI-based technology, which is designed to address the legal industry’s specific needs.