How transcription makes arbitration more efficient.

This article is to highlight one of the most ignored efficiency improvement methods of an arbitration process. If you are from the field of alternative dispute resolution, skip the next couple of paragraphs and jump straight in to read about transcription.

Arbitration is an alternative dispute resolution process where a neutral or a group of neutrals called arbitrators decide on a matter between two disputing parties who have chosen to address their dispute out of court. Arbitration is typically an expensive affair due to high costs of the arbitrators, lawyers fees, travel, accommodation and the fees of conducting the arbitration hearings.

Arbitration hearings typically happen in arbitration institutions or a chosen neutral venue like a boardroom in an upscale hotel. I am talking about the pre-Covid era where digital transformation of legal proceedings was still a research subject. Today most hearings happen online over video conferencing platforms, some tailor made for these processes. This has significantly reduced the costs of venue, travel and accommodation, thereby bringing down the overall dispute resolution costs. Along with cost reduction, it has also made ancillary services like transcription accessible and affordable.

What is Transcription?

Transcription, in the context of legal arbitration is a verbatim record of everything that is said during the course of a hearing. In an arbitration hearing where transcription is used, everything that is said goes on record. In cases where a dictation based stenographer service is used, only the information that the tribunal and the counsels dictate to the stenographer goes on record. I will limit this article to the former.

How are transcripts used?

Post the arbitration hearing, transcripts are referred to by the arbitrators to decide on the matter. This helps the arbitral tribunal to not make any false assumptions while they are drafting an award and instead, rely on the statements made by the parties, counsels and the witnesses that are accurately recorded in the transcript. Sometimes, the transcripts are also cited in the award. A typical 1 day hearing will generate a transcript of about 120 pages. That is a lot of information to just remember or make notes. Hence transcription becomes a very useful reference.

Real time transcription is where the spoken words and sentences appear on a screen in real time. This is helpful when the parties and the arbitrators want to refer to the transcripts while the hearing is in progress. This ensures that everyone is always on the same page and if someone missed any segment during the hearing, they can refer to the transcript and get back. It also helps in situations when a contradictory statement is made and someone wants to verify what was said before. A good real time transcript will have clear timestamps, speaker information and nearly accurate transcript for a user to refer while the hearing is ongoing.

When a hearing spans across multiple days, the parties have very little time to prepare for the next day based on the discussions on a particular day. An accurate transcript made available within a few hours after the conclusion of a day’s proceedings becomes especially valuable for this process. In the absence of a transcript, lead counsel would refer to notes made by them and their team of lawyers, spend a few hours trying to ensure that nothing is missed and then make further deductions based on them.

In cases where hearings are not happening on consecutive days, transcripts are usually procured by the parties a few days after the hearing is done. This reduces costs of transcription further.

As everything that is said during an arbitration hearing goes on record when transcription is used, the parties and the tribunal tend to be a lot more focused and minimise any inconsequential conversations. This further adds to the efficiency of a hearing.

How is transcription done?

A legal transcriber typically sits in the hearing room and types out everything they hear. Some use a stenographer machine (watch the video), also called a shorthand machine and others who are not trained with it use a regular qwerty keyboard. Stenographers using the shorthand machine undergo a rigorous training program before they are certified to offer their services and they are typically very expensive. Regular typists are affordable, but they are limited by their speed of typing and listening skills. A good transcriber, to be effective for arbitration proceedings, needs to be fast (at least 150 words per minute), accurate, alert throughout the hearing, have awareness of what is being discussed and be able to grasp accents of various speakers. Hence it is a very intense job and heavily dependent on the person doing it.

Transcription using technology

With the advent of machine learning and various Natural Language Processing (NLP)algorithms, speech to text has evolved leaps and bounds. There are plenty of services emerging that employ this technology to offer real time and delayed transcription for arbitration. An advanced speech to text engine offers more than 90% accuracy and the remaining accuracy can be achieved by a manual overlay. The advantages of technology based legal transcription are as follows:

  • No dictation and waiting for the steno or typist to complete typing: This is a big efficiency improvement over the current method where the pace of a hearing depends to a large extent on the typist and in most of these cases, the typists are dictated.
  • No losing of the flow of an argument or an examination: During an argument, any interruption could potentially derail the train of thoughts. During an argument, the speed of speech also increases which makes it difficult for a human transcriber to grasp everything and they end up interrupting the flow by requesting to repeat a word or a sentence.
  • Availability of an audio or video recording for further verification: A good transcription service provider provides a transcript in digital format along with a recording of the hearing. The software is designed to help the users click a segment of the transcript and hear only that part of the audio.
  • Way cheaper than stenographer: Technology should bring costs down and legal transcription is no exception. As these services come at a fraction of a cost of stenographers, legal transcription is expected to be availed in more arbitration hearings and hence will contribute towards improving the efficiency of dispute resolution at scale.
  • Improves with time: The concept of Machine Learning means that the more we use it the better it gets. Hence the speech to text engines are known to become more accurate with time and hence become even more valuable than they already are.

Although technology based legal transcription offers a great deal of advantages, it comes with its own challenges. Some of them are:

  • It needs a very good technology and design team to work in tandem to build a useful solution.
  • There will be a manual overlay after the speech to text engine has done its job to improve the accuracy further. The person correcting the transcript needs to be cost effective and still have most of the characteristics of a real time stenographer as described before.
  • Transcription of multiple languages in the same session is not efficiently possible.
  • The data security and privacy norms must be strictly adhered with due to the sensitive nature of the data.

Should a simple hearing need transcription?

In most simple cases, hearings last a few hours and there is a long gap between two successive hearings. In such cases, transcription may seem a bit of an overkill. The parties can choose to use a stenographer to record the key points and that will mostly be sufficient. In the absence of a stenographer, a recording of the hearing can be transcribed and saved for reference. Note that a delayed transcript costs a lot less than a real time transcript or same day delivery transcript.

After conducting several complex arbitration hearings on CORD, we can markedly see the efficiency transcription brings into the process. A recent hearing that took place on the CORD platform was planned for 8 days and was eventually completed in 6.5 days largely due to the real time transcription and robust virtual hearing methods. 15 participants saving 1.5 days resulted in a huge cost saving that will benefit every stakeholder of an arbitration hearing.

About the author: Badarivishal Kinhal is the co-founder of CORD. He lives in the intersection of law, technology and art.