Privacy Policy

Who uses this service

Accounts are created by an administrator, not by self-service sign-up. Every user is a court reporter or a member of a reporting firm that has an existing relationship with us. The service is not directed at children and is not intended for personal use.

What we hold

Account information

Your email address, an optional display name, and your password — stored only as a scrypt hash, never in a form we can read. Signing in sets a cookie that identifies your session for seven days.

Recordings

Audio you upload, record in the iPhone app, or capture through a meeting bot. Audio is stored in Amazon S3 and is retained until you or an administrator deletes the job.

Transcripts

Machine transcripts, your edits to them, speaker names, and the hearing details you enter — name, date, and location.

Job and usage records

Which jobs exist, who owns them, who they are shared with, when they were processed, and how much transcription was run. We do not use advertising identifiers, analytics SDKs, or any tracking technology, and the apps contain no third-party SDKs at all.

The iPhone app

Truth Capture records to the phone and transcribes on the phone, using Apple’s on-device speech recognition. That preliminary transcript is produced without a network connection and the audio is not sent anywhere to create it — it never leaves the device during recording, which is the point of the feature.

When you choose to upload a recording, the audio and the on-device draft are sent to the same service described above, and the recording is transcribed again on the server with speaker separation. The local copy stays on your phone until you remove it; the app never deletes your audio on its own.

The app asks for microphone access to record and for speech recognition access to transcribe. It requests no other permissions, and it collects no location, contact, or device-advertising data.

Who else processes this material

We use a small number of service providers to run the pipeline. They process material on our behalf, and none of them are permitted to use it for their own purposes.

Amazon Web Services

Stores audio, transcripts, and account records, and performs the server-side speech recognition that produces the transcript with speaker separation.

Recall.ai

Used only when you send a bot to a video meeting. Recall joins the meeting, records it, and returns a transcript; it uses AssemblyAI to distinguish speakers who share a single microphone. If you never dispatch a meeting bot, no material reaches them.

A large-language-model provider

Used only when you ask the editor to suggest who the speakers are. That feature sends transcript text, never audio, and only for the transcript you are working on. If you do not use the feature, nothing is sent.

If you report a problem from inside the app, the report is stored with your account and may include a screenshot of your screen, which can contain transcript text. Screenshots are only ever captured when you file a report.

Operational notifications

So that we notice problems quickly, the service notifies its operator when a job is created, when a meeting recording arrives, when an account is created, and when someone files a problem report. These messages carry your name, email address, and the job’s title and identifiers. They do not carry recordings or transcript text.

Who can see your work

A job is visible to the person who created it, to anyone that person shares it with — with either view or edit permission — and to administrators of the service, who need access for support and billing. Nobody else. We do not sell material, share it for advertising, or use recordings or transcripts to train machine-learning models.

Keeping it and deleting it

Recordings and transcripts are kept until deleted, because a transcript is a record that has to remain available for as long as the proceeding requires. Deleting a job removes its audio, transcript, and metadata from the service. To close an account and remove everything it owns, write to us at the address below and we will action it.

Security

Traffic is encrypted in transit with TLS. Passwords are stored as scrypt hashes. Uploads go directly from your browser or phone to storage using short-lived signed URLs, and the credentials that can read media are held only by the background worker, not by the web application. No system is perfect, and we do not claim otherwise.

Your responsibility for recording

Recording law varies by jurisdiction and by proceeding. You are responsible for having the authority and any consent needed to record what you record. This tool does not establish that authority for you.

Changes

If this policy changes materially, we will update the date at the top of this page and tell account holders directly.

Contact

Questions about this policy, or a request to remove your data, go to support@suranyi.systems.