It is straightforward to record people talking during a phone call, at a meeting or even in court. The challenge is turning that recording into text.

A person such as a legal secretary can sit through the recorded audio and manually transcribe, knowing the names of the speakers and recognising each person’s voice. This is a costly and time consuming process. A standard file note procedure can usually achieve the same objective more efficiently.

The benefit of turning voice into text is that text is a format that is part of every law firm’s existing processes. It is also a format required by courts and the legal system generally. Both humans and technology are adept at reading and searching text.

Voice recording is growing exponentially in other industries and audio files are becoming a necessary element in eDiscovery, as we reported here.

There may come a time when audio becomes a standard part of record keeping in law firms, especially as technology develops for searching and extracting from audio files. In the meantime, technology is being applied to transcription.

Advances in transcription technology

A technology race is currently taking place to achieve the ideal of accurate automated transcription, i.e. voice to text, with minimal or no human intervention.

The technology race is being stimulated by advances in affordable voice capture tools such as Google Home and Amazon Echo. See our recent article on a proposed application of Google Home in the legal sector here. The combination of advances in hardware and the accessibility of artificial intelligence and machine learning is leading to new entrepreneurial activity in this field.

Two such startups are Tetra and Tali.

Tetra is a voice transcription startup that recently raised US$1.5 million in seed funding. Read more at TechCrunch.

Tali is a startup that has targeted the legal sector with what it calls “conversational time tracking”. Integrating conversational time tracking with legal practice management systems is part of this company’s plans.



While voice-to-text transcription is not new, it is the application of artificial intelligence and machine learning that has created new possibilities for streamlining record keeping and transforming age-old practices.

© 2017 Legal Practice Intelligence