Privacy Policy

Effective September 12, 2026. SpeakMyLingo is designed to process only the information needed for the capture mode and features you choose.

Recording and consent

SpeakMyLingo starts microphone capture only after the user confirms that participants have been informed. Recording and transcription laws vary by location. The person operating SpeakMyLingo is responsible for obtaining any consent required where the conversation occurs.

Meeting content

When you use an online transcription, translation, speech, or meeting-intelligence feature, the audio or text required for that request is sent to the selected AI provider. SpeakMyLingo does not send meeting audio, transcript text, titles, names, keywords, provider keys, viewer keys, or private notes to product analytics.

OpenAI states that API data is not used to train its models by default. Its current data-controls table lists up to 30 days of abuse-monitoring retention for the Realtime API unless the applicable OpenAI organization has approved data-retention controls such as Zero Data Retention. SpeakMyLingo therefore describes what it retains separately from what a processing provider may temporarily retain. See OpenAI API data controls.

Your privacy mode

Encrypted audio + transcript keeps encrypted audio and the transcript on your device. Transcript only keeps the transcript but no durable audio recording. Live only means SpeakMyLingo keeps neither audio nor transcript after the session ends and disables export and outside-app viewing; live audio still passes to the disclosed AI processor. Optional post-meeting refinement uses retained content only when you request it.

Storage and deletion

Retained meeting audio is encrypted with a per-meeting key protected by the iPhone Keychain. Deleting a retained meeting destroys its key and removes its files. A delete-after-processing option removes retained audio after successful processing; a failed processing attempt keeps encrypted data available for explicit recovery rather than silently losing it.

Live viewer

The optional laptop viewer receives encrypted transcript envelopes. The decryption key remains after the URL fragment and is not sent to the SpeakMyLingo server. Anyone who receives the complete viewer link may be able to read that live session, so share it only with intended participants.

Analytics and purchases

Optional in-app product analytics are disabled until you opt in. Autocapture, session replay, screen capture, surveys and advertising tracking are disabled. Apple and RevenueCat process purchases needed to provide minute packs. The backend records content-free purchase, refund, minute-grant and usage events in PostHog to operate and reconcile the service. These records use a pseudonymous installation account ID and contain no meeting content.

SpeakMyLingo requires no name, email address or login. An account identifier generated by the app is stored in the installation’s Keychain and associates purchases, service usage and opted-in product analytics with that installation. These records are pseudonymous, not fully anonymous. AI requests also use a derived safety identifier. Analytics person profiles are normally disabled; deletion may temporarily create a profile containing only the account ID so PostHog can erase associated events. Consumable minutes are not automatically transferable to another device.

To prevent repeated redemption of the one-time welcome offer, a new installation may send an ephemeral Apple DeviceCheck token to SpeakMyLingo and Apple. SpeakMyLingo stores only a pseudonymous one-way SHA-256 hash and claim time, never the raw token.

Operational diagnostics

SpeakMyLingo may send the app build, iOS version, and aggregate counts of crashes, hangs, CPU exceptions, and disk-write exceptions so reliability problems can be detected. These reports contain no audio, transcript, meeting title, participant, note, viewer link, or advertising identifier and are not used for tracking.

Service providers and transfers

Depending on the selected feature, data may be processed by Apple, OpenAI, RevenueCat, PostHog, Vercel, or Supabase under their respective terms and regional infrastructure. This can involve processing outside the country where the conversation occurs.

Your choices

You can select Live only, disable analytics, delete retained meetings, or stop capture at any time. Read the recording-consent guide. SpeakMyLingo does not ask you for an AI-provider key. For support, email sayspan7@gmail.com. Never send real meeting audio, transcripts, access codes, credentials, or viewer links in a support request.