The short version
- Live transcription sends audio to a cloud speech provider.
- Transcript text is uploaded to our database when you choose Save.
- Saved transcripts stay until deleted; inactive accounts do not expire.
- The mobile app uses PostHog and Google Analytics for Firebase.
- This website’s preview uses sample text and does not record you.
- Delete your account in Settings, or request account or data deletion by emailing sharmaga440@gmail.com. View deletion instructions.
1. Who we are & scope
This policy covers LiveScribe, its transcription service, and this website. LiveScribe is offered under the trading name TreasureTask, from Ontario, Canada, and is intended for adults worldwide, subject to availability and applicable law. “We”, “us”, and “our” mean the operator doing business as TreasureTask, responsible for deciding how personal information is used for these services.
Operator: Garvit Sharma, trading as TreasureTask
Location: Ontario, Canada
Privacy contact & support: sharmaga440@gmail.com
Audio may contain information about the app user and other people nearby. If someone else uses the app to transcribe your speech, we receive that information from their device. You may contact us about your information even if you do not have an account. Separate policies apply to services you visit through external links.
2. Information we process
| Information | Source & purpose |
|---|---|
| Account information | Apple or Google sign-in, through Supabase authentication, provides an account identifier, email (which may be an Apple relay address), and available profile name. We use these to sign you in, secure your account, and associate your transcripts and purchases with you. We do not receive your Apple or Google password. |
| Audio & text | Microphone audio is streamed when you start transcription. Speech recognition produces draft and finalized text. Transcript drafts are stored on your device. When you choose Save, transcript text and any edits or text included in that transcript are uploaded to our database so you can revisit them. Text entered only in the reply display is not automatically saved as a cloud transcript. |
| Session & usage information | Session identifiers, selected transcription model, timestamps, recording duration, session status, and whether the introductory recording has been used help run the service and apply access limits. |
| Purchase & balance information | RevenueCat and the relevant app store process purchase identifiers, products, subscription status, and related transaction details. Our backend stores purchase references, subscription allowances, purchased extra hours, and usage balances to provide paid access and verify eligible restored purchases. We do not collect your full payment card details. |
| Device & technical information | Services receiving a connection can process IP address, device/app version, request times, and operational errors to deliver, protect, and troubleshoot the service. Device settings and authentication tokens are stored locally as needed. |
| App analytics | PostHog and Google Analytics for Firebase receive app interactions, screen views, sign-in and purchase events, recording duration, account or app-instance identifiers, and device/app information to help us understand usage and improve reliability. Our app analytics events do not include microphone audio, transcript contents, typed replies, or feedback message contents. |
| Support messages | In-app feedback is stored with your account identifier, message, and submission time. Emails and privacy requests are used to respond, verify requests, and keep necessary records. Avoid sending unnecessary audio or sensitive transcripts. |
Audio and transcripts may reveal sensitive information, including health, beliefs, or other people’s private details. Speech-to-text conversion in the app does not create a voiceprint to identify a person. Do not submit sensitive information unless you have the required permissions and a lawful basis.
3. How transcription works
After microphone permission is granted and you start a session, the app sends audio through our server to the selected speech recognition provider: AssemblyAI or Deepgram. Their results pass through our server back to the app. Draft transcript text stays in device storage until you choose Save; choosing Save uploads that text to our database. Session metadata, titles you set, recording duration, and usage records can be stored on the server before you save the transcript text.
Audio is streamed through temporary buffers; our application database does not store raw audio recordings. Speech providers and infrastructure services process the data needed to deliver the service and may retain operational records under their applicable terms. A connection is needed for live transcription. Keeping an unsaved transcript on your device does not mean its audio or transcription was processed only on the device.
Pause or stop transcription to stop sending new microphone audio. You can revoke microphone permission in your device settings. Revoking permission does not erase information already processed.
Provider use and model improvement: we do not operate a model-training feature in LiveScribe. Speech-provider handling depends on the applicable service terms and training controls. The providers distinguish conversation content from usage records:
- AssemblyAI streaming: with model training opted out, AssemblyAI documents zero retention of audio and transcripts. It still retains metadata for logging and billing, without a fixed period stated in its production-retention guidance.
- Deepgram streaming: when its model-improvement opt-out applies, audio and transcripts are retained only as needed to process the request. Request metadata and usage logs remain retrievable for 90 days, and summarized usage is kept longer. Without the opt-out, content is retained for model improvement. See Deepgram’s data-handling documentation.
4. Purposes & legal grounds
We use information to provide transcription and account features, manage subscriptions, understand app usage, respond to requests, prevent abuse, secure the service, and meet applicable legal obligations. We do not use transcripts to make decisions with legal or similarly significant effects about you.
Where the EU or UK GDPR applies, the proposed grounds are: performance of our agreement for account services, requested transcription, and subscription access; legitimate interests for proportionate security, troubleshooting, and fraud prevention; legal obligations for required records and lawful requests; and consent where required for optional processing or sensitive information. Our security interest is keeping accounts, systems, and people’s information safe.
Contract necessity does not automatically cover everyone whose speech is captured, and microphone permission is not everyone’s legal consent. Leaving names out of analytics does not anonymize the audio being transcribed: a voice or details in a conversation may identify someone, and speech can contain sensitive information. Before transcribing other people, let them know that their speech will be processed by a cloud transcription service and obtain any permissions required by applicable law.
Providing account and session information is necessary for the corresponding features; you can choose not to provide it, but those features will be unavailable. Where processing relies on consent, you may withdraw it without affecting earlier lawful processing.
5. Who receives information
Information is disclosed to services needed to operate the features you use:
- AssemblyAI or Deepgram: audio and session parameters for the selected speech-to-text service. See AssemblyAI’s privacy information and Deepgram’s privacy information.
- Supabase: account authentication and database storage for account records, saved transcripts, session and usage information, feedback, and billing records. Our project is hosted in the United States, in region us-west-2.
- DigitalOcean: hosting in New York, United States, for our application server, which handles authenticated requests and relays audio and transcription results. It also processes connection and operational information needed to run the backend.
- PostHog and Google Analytics for Firebase: app usage analytics as described in the analytics section below, including identifiers and interaction events, but not the contents of your conversations.
- RevenueCat, Apple, and Google: purchase validation, subscription access, sign-in where selected, and app distribution. Store payment processing is governed by the applicable store’s terms and privacy notice.
- Expo: app delivery and update infrastructure, which may receive technical information when checking for or downloading updates.
- Vercel: hosting for this website, including connection and security information needed to serve pages. Email providers also process messages you send to our privacy and support address.
We may also disclose information when legally required, to investigate abuse or protect rights and safety, to professional advisers bound by confidentiality, or in a business transfer subject to applicable safeguards and notice. Sharing with a processor must be limited to its agreed purpose; a vendor’s public privacy policy does not replace our responsibility to assess and contract with it.
The app and website do not include features for selling your conversations or displaying third-party advertising. We use app analytics to understand usage and improve the service. Analytics provider settings and any advertising-related integrations must be assessed separately from the events explicitly sent by the app.
6. Retention & deletion
Delete your account in the app: open Settings, select Delete account, and confirm. You can also request deletion by emailing sharmaga440@gmail.com with the subject “LiveScribe account deletion”. Send the request from the email linked to your app account where possible. You do not need to install or sign in to the app to use this email option. We may ask you to verify ownership; never send your password.
Account deletion removes your sign-in account, profile, saved transcripts and included text, session records, in-app feedback, and account-linked billing and usage balances from our active database. The in-app flow also clears that account’s transcript drafts on the device used to delete it. It does not remotely erase copies on other devices, copies you exported, or every record held separately by a provider. You can also email us to request deletion of particular data without closing your account. For emailed requests, we will confirm completion and explain any retention exceptions. See Account & data requests for instructions.
Saved transcripts and accounts: saved transcripts and associated session records remain until you delete them or your account. Inactive accounts do not automatically expire. Account records, in-app feedback, and account-linked usage and balance records remain while the account exists, unless deleted in response to a separate request. Deleting a transcript does not reset used transcription time or the introductory-session usage flag.
Deletion timing for records we control: we delete the requested records from our active systems immediately once the request has been verified, without an additional retention period. In-app account deletion removes the active database records as part of the deletion request. For emailed requests, verification and handling happen before deletion; sending an email does not itself erase data.
Provider records and backups: deleting active records does not erase every provider-held copy. Speech-provider metadata, analytics, store transactions, and backups are subject to separate provider processes. Our account-deletion flow does not automatically submit deletion requests to PostHog, Google Analytics, or RevenueCat. Contact us about associated provider data. Provider processing can take time; for example, PostHog processes person-data deletion asynchronously.
Uninstalling the app or signing out does not delete your account or cloud transcripts. Account deletion also does not cancel an app-store subscription. Deleting your account removes its unused transcription balance; store purchase restoration is separate from restoring deleted account data. See the purchase and restoration terms.
7. Device storage, website & app analytics
This website does not add advertising pixels, analytics SDKs, nonessential cookies, or browser storage. Its interactive preview uses preset text in memory and never accesses your microphone. Fonts are served with the site rather than requested from a third-party font service.
Hosting infrastructure may process necessary connection or security information. The mobile app stores authentication credentials, preferences, transcript drafts, and analytics identifiers or queued events using device storage. Its processing differs from this informational website.
Mobile app analytics: we use PostHog and Google Analytics for Firebase to understand how people use LiveScribe and to improve reliability. Events include screen views, button interactions, app activity, settings changes, sign-in outcomes, recording and saving events, durations, and purchase or restoration outcomes. Analytics can associate these events with your account ID or an app-instance identifier. SDKs may also process device/app details and connection information; Google Analytics may derive approximate location from an IP address.
We do not include microphone audio, transcript text or titles, typed replies, feedback contents, names, or email addresses in the app analytics events we send. PostHog session replay, automatic error capture, and IP-based geolocation enrichment are disabled in our app configuration. These settings do not mean the analytics services receive no connection information. See PostHog’s privacy notice and Google’s privacy notice.
Analytics retention: PostHog’s published event-retention policy provides a one-year event-retention window for our project. This window is not a substitute for requesting deletion of a person and their events. Google Analytics user-level retention depends on the property’s configuration, with two- or fourteen-month settings. New activity can reset the retention period for a user identifier, and standard aggregated reports are outside that setting. See Google’s retention guidance.
The current app does not provide an in-app analytics toggle. Contact our privacy address to make a request about analytics data associated with your account. Signing out does not by itself stop app analytics. Where consent or other collection controls are legally required, this policy is not a substitute for those controls.
8. International processing
We operate from Ontario, Canada. Our Supabase project stores information in the United States (us-west-2). Our DigitalOcean application server is in New York, United States. The app’s PostHog configuration points to its US service. Our speech connections use the providers’ default global endpoints, rather than endpoints restricted to one region. Speech providers and other services may process information in additional countries, including outside your country of residence. Protection and government-access rules may differ.
AssemblyAI includes its Data Processing Addendum in its standard service terms; its published agreement includes EU standard contractual clauses and UK transfer provisions. This does not establish the agreements or safeguards applicable to every other provider.
Contact our privacy address for information about processing locations and applicable safeguards. Worldwide availability does not remove local privacy rights or recording rules. We do not claim that all processing stays in Canada or on your device.
9. Your rights & choices
Depending on your location and the law that applies, you may request access, a copy or portable export, correction, deletion, restriction, withdrawal of consent, and information about processing and recipients.
Right to object: where processing relies on legitimate interests, you may object based on your circumstances. You may also object to direct marketing; this website does not operate a marketing mailing list.
Use the contact details above to make a request, appeal a refusal where applicable, or authorize someone to act for you. We may ask for proportionate identity verification, but not unnecessary sensitive information. We will respond within the legally required period, explain permitted extensions or refusals, and not discriminate against you for exercising protected rights.
- EEA, UK, and Switzerland: you may raise concerns with your local supervisory authority. UK residents can contact the ICO; EEA residents can find their national authority.
- Canada: applicable federal or provincial law may provide access, correction, consent, and complaint rights. You can contact the Office of the Privacy Commissioner of Canada or the relevant provincial authority.
- California and other US states: where their laws apply to this operator, additional rights can include knowing categories and disclosures, correcting or deleting data, opting out of sale, sharing, targeted advertising or certain profiling, limiting certain sensitive-data uses, and appealing decisions. See the California Attorney General’s guidance for the scope of California rights.
- Other regions: local rights, including those available under Brazil’s LGPD or other applicable privacy laws, are preserved. Contact us to exercise a right or raise a concern.
People whose speech was transcribed can also make requests. Provide only enough detail to locate the relevant session. You do not have to contact us before complaining to a competent regulator.
10. Security & age eligibility
Production connections are intended to use HTTPS and secure WebSockets, with account authentication and access checks for transcripts. These measures reduce risk but cannot guarantee absolute security. Protect your device and sign-in account and avoid transcribing highly sensitive conversations unless the service has been assessed for that use.
LiveScribe is intended for adults aged 18 and over and is not directed to children. People under 18 are not eligible to use the service. If you believe a child has submitted personal information, contact us so we can investigate and delete information where required.
11. Changes & contact
We will update this page when our practices change and display the revised date. Material changes will receive appropriate notice, and we will obtain consent when legally required before using information for a new purpose. A change to the policy does not itself authorize new processing.
Questions about privacy, access to this policy in another format, and privacy requests should go to the contact shown in section 1. The revision date and effective date for this update are shown at the top of this page.