Legal transcription services – Advanced speech recognition

As digital dictation has developed, so too has speech recognition technology.  Most firms now dispatch dictated digital recordings to the traditional typing pool or outsource, but are increasingly considering a speech recognition service.   Is the latter a commercial solution being adopted widely within the industry?  Is speech recognition uptake where it should be? Accuro discusses further in a Q/A below:

A few facts (according to Lawyer Monthly) – showing law firms maybe willing to embrace speech recognition technology:

  1. 82% of legal firms aim to invest in speech recognition technology going forward.
  2. 7% of legal professionals reportedly spent between 2 and 4 hours a day typing.
  3. Only 19% made use of internal typists, and 5% used external transcription services on a regular basis.
  4. 62% of legal professionals who were not currently using speech recognition technology said that they would in future.


Question and answer session on speech recognition and its current viability in the legal sector.

Question – are firms willing to adopt speech recognition technology?

I think there is a general willingness to do so as the cost and time savings are hard to ignore. I think there is a viability issue in a commercial environment which is preventing a quantum shift over to speech recognition technology.

Question – can you explain ‘viability issue’?

This is a complicated area and depends on user by user or firm by firm. Simply put, if the expectation is that audio (dictation) converted by a speech recognition engine will produce an intelligible document with high standards of accurately typed text and correct formatting, then users will be disappointed. A significant amount of work is required to proofread and correct the text, especially for external facing documentation, emails, attachments, etc.  

If, however, the user wants a quick speech-to-text conversion and is happy to work with the result then a stand-a-lone speech engine may suffice, if editing time is considered on the costings. At Accuro, this is called ‘rough draft’.

Question – what do you consider a solution for firms currently?

If we consider a typical legal customer at Accuro, a portion of their audio may be simple file notes that are not for general consumption and quickly converting speech into text may be sufficient. Another portion may be detailed conference video calls, with multiple attendees and with copies being distributed to all, this would be better suited to human legal transcription service.

What is evident is users may require a simple one-stop-shop for their dictation. Accuro provides the choices and allow uploads to different services. Simple, single voice files can go to speech recognition either with or without editing, and external work requiring high accuracy going to the human transcription service.  

Accuro has a simple user interface that allows uploads to be routed to the best solution. Choosing the right speech-to-text service significantly reduces the cost to the customer but allows them to receive a quality of service that they need.

Question – what is an enhanced speech recognition product?

Speech recognition will continue to improve and it’s important that firms embrace the technology, the more people using it, the better it becomes. Accuro’s ‘enhanced speech recognition’ product comes with a professional proofread which maximises quality. The product is much cheaper than outsourced transcription and by eliminating errors of a ‘rough draft’, gets the business involved in AI solutions without the risk of low quality.

Question – is the solution complicated to install and use?

It’s very simple and requires no software installation. Accuro provides secure access to a web-based platform and customers can easily route their dictations by dragging and dropping or browsing.

This web-based transcription platform provides a dashboard with work in progress, performance and easy access to completed files/transcripts.

Accuro is also integrated with most dictation workflow platforms such as BigHand, Lexacom, Crescendo etc. and again, requires very minimal setup time.

Question – how would you summarise Accuro’s legal product?

The legal space is made up of many profiles of firms. Some may have tried to implement speech recognition with an unsuccessful take up, others may rely on internal typists with little technology. Whatever the profile this range of products; transcription, enhanced speech recognition and rough draft allows any firm to access secure speech-to-text products which maximises quality to provide a real reduction in costs. 

What is ‘enhanced’ about Accuro’s enhanced speech recognition?

The cost and time savings of introducing speech recognition into a commercial environment are persuasive. So why has speech recognition not swept the board?

Accuro spent 12 months testing a host of professional speech recognition platforms encompassing ‘real world’ audio including single voice (file notes etc.) and multi-voice (calls with Counsel etc.) and on varying platforms including Teams/Zoom, telephone calls and office-based dictation. Our conclusion was speech recognition is viable if your end user has time to complete a detailed proofread and reformat the documents. If this ‘edit’ is not completed the documents fall significantly short with an error rate wildly varying between 40% and 80% on a word count and layout perspective. 

Enhanced speech recognition is a professional proofread of the document with any errors corrected.  This unique service allows law firms to significantly reduce dictation costs from the traditional human transcription service. Most importantly, users can choose to upload audio to this enhanced speech recognition service, speech recognition only or outsourced transcription, allowing the end-user to match their requirements at a ‘one stop shop’ for speech-to-text services. 



The most accurate speech-to-text service. A direct replacement or help for in-house staff, completed by UK-based, skilled transcribers and professionally proofread to achieve maximum quality.


Rough draft
(speech recognition)

A really quick way to get your audio file returned in minutes. Ideal for users who want a cost-efficient rough draft to get started. Available in both single-voice and multi-voice formats.


(speech recognition)

A cost-effective AI solution with the benefit of being professionally proofread. Available in both single-voice and multi-voice formats.

As well as our high-quality professional and legal transcription services, Accuro provides excellent medical transcription services, academic transcription services, property transcription services, captioning servicestranslation services and Zoom & Teams transcription services.

For more information, contact us or get started by setting up a free account by clicking here.

Share this post