Parakeety for lawyers and solicitors

Lawyers spend half the day writing into different windows: an attendance note straight after a client meeting, a draft of a witness statement, a long email to the other side, a section of a brief, a chronology that has to be done before five o'clock. The bottleneck is generating words quickly enough that the day does not slip. Cloud dictation tools wager on giving you polished prose at the cost of audio leaving the Mac. Parakeety wagers the other way: raw transcript at the cursor, in around 0.2 seconds for a ten-minute dictation, with nothing leaving the machine.

It pastes wherever the cursor is. That includes Clio, Actionstep, LEAP, ProClaim, MyCase, PracticePanther, Smokeball, plus Word, Outlook, browser-based document tools, and Notes. Parakeety does not integrate with the case-management system specifically. The integration is that the operating system already knows where your cursor is, and Parakeety pastes the transcript there.

Client confidentiality and privilege are the load-bearing duties in legal practice. The SRA Code of Conduct and the Bar Standards Board's handbook set out the UK position; ABA Model Rule 1.6 and the equivalent state rules set out the US position. Privilege protects communications and work product from compelled disclosure, but only where the privileged material has not been waived. Voluntarily disclosing audio of privileged material to a third-party processor is at minimum something to write into the firm's data-processing register, and at worst a defensible-waiver argument waiting to happen if the disclosure is contested.

Cloud dictation services typically handle this by writing a contract: a data-processing agreement under UK GDPR, a Business Associate Agreement under HIPAA where medical matter is in scope, an information-security questionnaire signed at procurement. That works for many firms. The cleaner answer for the same problem is to not transmit the audio in the first place. Parakeety processes audio on the Apple Neural Engine, in memory, then discards the buffer. There is no third-party processor in the data-flow diagram because there is no network call at transcription time. That is the architectural answer; whether your firm's risk team needs it or accepts the contractual one is a separate question, but it is a useful answer to have. The same distinction maps directly onto data-processing reviews and SRA-handbook conversations; the full version is in HIPAA and dictation: architectural vs contractual privacy, written for the healthcare audience but with a section on legal professional privilege that reads across to UK and US firms. Local dictation for clinicians restates the same argument from the medical side, where HIPAA does the heavy lifting privilege does here.

• Each license covers one Mac. A laptop per fee earner, or a shared dictation Mac in the secretarial pool, both shapes are fine; the model is one license per machine.
• Audio is captured to a memory buffer while the key is held. As soon as you release, the buffer goes through the speech model, the transcript pastes at the cursor, and the buffer is discarded. Nothing is written to disk. Nothing is logged.
• The model auto-detects the language across 25 European languages, which matters for cross-border work and for dictating quoted material from a foreign-language source. Local dictation for translators covers the translation-side variant of that workflow.
• The push-to-talk key is currently the section key (§, below Esc). A user-configurable shortcut is on the roadmap, which matters if you have already mapped that key in your case-management software.
• First launch downloads the speech model (around 600 MB) and asks for microphone and accessibility access. After that the app lives in the menu bar and runs offline.

Worth being clear about. Parakeety is not a court-grade transcription tool. It does not produce certified verbatim transcripts and is not built for hearings, depositions or recorded interviews. It is push-to-talk for the moments when you are writing the attendance note, the email, the section of the brief; not a tool that captures audio of a meeting and produces a transcript. If your need is for verbatim transcription of a recorded session, or for a tool with a contracted data-processing agreement in place with the vendor, this is not the right shape of answer. For more on that distinction we covered the difference between live push-to-talk dictation and recorded transcription in Parakeety vs Wispr Flow.

Parakeety is a Mac menu-bar app. Hold the section key, talk, release; your words paste at the cursor in whichever app you were typing into. Audio never leaves the machine. There is a free 7-day trial with no card required. After that it is $30 once.

Try Parakeety free →