Privacy Policy

Account information. Your email address, display name, optional profile photo, the languages you speak or are learning, your interface language, colour theme, and home currency.

Voice and conversation content. When you use the translator, we process audio from your microphone, the resulting transcript, and the translated text so we can show and speak the translation. Conversations you keep are stored in your history so you can revisit them. Contacts you add and messages exchanged with them are stored so both participants can see the conversation.

Voice Key (biometric voice data). If you choose to set up a Voice Key, we create a mathematical representation (a voice embedding) of how you sound so the app can filter out other speakers. This is optional, is only created when you explicitly record an enrolment sample, is stored privately against your account, and can be deleted by you at any time from the Voice Key screen.

Technical and usage data. Device type, browser, approximate location derived from your IP address or (with your permission) device location for local currency, app version, page views, feature usage, error reports, and performance timings.

• To provide the core service: transcription, translation, speech playback, pairing, messaging, and history.
• To personalise your experience: preferred languages, themes, currency, and learning material.
• To send notifications you have opted into, such as knock requests.
• To keep the service secure, prevent abuse, and diagnose faults.
• To improve the accuracy and quality of speech recognition, translation, and speech synthesis — see section 3.
• To comply with legal obligations.

Speech recognition and translation improve when they are trained and evaluated on real speech from real conversations, across accents, dialects, and noisy environments. Unless you opt out, an anonymized subset of audio snippets, transcripts, and translations may be used to develop and improve our models and the third-party models we rely on.

Before any such use:

• Your account identifiers — name, email, user id, and contact details — are removed and replaced with a random reference so the sample is not linked back to your account.
• Samples are handled under strict access controls and confidentiality obligations.
• A small number of samples may be reviewed by trained annotators or automated systems to verify quality; reviewers see only the anonymized snippet.
• Voice Key embeddings are never used for model development.
• Payment details are never used for model development.

Your choice. You can turn this off at any time in Profile → Data & privacy. When the opt-out is on, we stop using your data for service improvement from that point forward, and it does not affect your ability to use any feature. Anonymized samples already incorporated into a trained model cannot be individually withdrawn, because they are no longer identifiable or separable.

We use Google Analytics (provided by Google LLC) to understand how the app is used — for example which screens are visited, which features are used, and where errors occur. Google Analytics sets cookies or similar identifiers and collects your IP address (which Google truncates/anonymizes), device and browser information, and interaction events. This data is processed for statistical analysis and product improvement, and may be transferred to and processed on servers outside your country, including the United States, under appropriate transfer safeguards.

We do not send your conversation content, transcripts, audio, or Voice Key data to Google Analytics. You can opt out of Google Analytics across all sites using Google's browser add-on, or by blocking analytics cookies in your browser.

We share data only as needed with providers who process it on our behalf under contractual confidentiality and security obligations: cloud hosting and database services, speech-to-text and translation model providers, text-to-speech providers, push notification delivery, email delivery, currency exchange-rate data, and analytics. We do not sell your personal information, and we do not share it for cross-context behavioural advertising.

• Account and profile data: kept while your account is active.
• Conversation history and messages: kept until you delete them or delete your account.
• Voice Key: kept until you delete it or delete your account.
• Guest (unregistered) sessions: automatically deleted after a short retention window.
• Anonymized samples used for service improvement: retained only as long as needed for model development and evaluation, and no longer linked to you.
• Analytics data: retained according to our Google Analytics retention settings.

Data is encrypted in transit and at rest. Access to conversation content is restricted by per-account access rules so that only participants of a conversation can read it, and internal access is limited to staff who need it to operate the service.

Depending on where you live, you may have the right to access, correct, export, or delete your personal data, to object to or restrict certain processing, and to withdraw consent. You can edit your profile and delete conversations, vocabulary, and your Voice Key from within the app, and you can request full account deletion by contacting us. We will not discriminate against you for exercising these rights.

PolyPort.Chat is not directed at children under 13 (or the minimum age in your country), and we do not knowingly collect their personal data. If you believe a child has provided us data, contact us and we will delete it.

We will post any changes to this policy on this page and update the date above. For privacy questions or requests, contact us at privacy@polyport.chat.