Speakwise Privacy Policy

Last updated: August 3, 2026

This policy explains what information Speakwise ("the app", "we", "us") collects when you use it, why we collect it, who else sees it, and what choices you have. Speakwise is an English-learning app that helps you practice speaking, fluency, and clarity.

1. Information we collect

Account information. When you create a Speakwise account (by email and password, Google Sign-In, Apple Sign-In, or phone number), we collect your name, email address, and/or phone number, whichever the method you chose provides. Passwords themselves are never stored by us — they're handled entirely by Firebase Authentication. During sign-up we ask you to confirm you are 13 or older (self-attestation only — we do not independently verify age).

Profile details you choose to add. Date of birth and profession, if you fill them in on the Edit Profile screen. These fields are optional.

Learning activity. Your CEFR level and progress within each level, which practice scenarios you've completed, your practice streak, minutes practiced per day/week, your learning goal (e.g. business, exam prep, accent), quiz/play history summaries, and whether you've completed the placement test.

Voice recordings and speech data. To score your speaking practice, Speakwise records your voice and, where applicable, converts speech to text on-device using your phone's (or browser's) built-in speech recognizer. You can pick an English recognition accent (United States, India, United Kingdom) in Settings → Voice & Audio. Scoring uses your transcript text and recording duration — text/STT-inferred clarity and fluency feedback, not true acoustic pronunciation analysis of raw audio.

Optional practice video (on by default on your device). You can also record camera video during Practice for your own body-language self-review (default on locally; turn off on the Practice prompt). Video stays on your device for the session and is cleaned up afterward unless you export a copy. Speakwise does not upload live Practice camera video as a cloud archive and does not use camera video for body-language grading.

Score a recording (optional upload). From Practice you may pick a short speaking video (MP4/WebM/MOV, size-capped) to score. The file stays in session memory only — Speakwise does not store that video in Firebase Storage. We extract audio on your device/browser, may briefly send that audio through our Cloud Function for cloud speech transcription, then score the transcript text with AI. Leaving the screen or refreshing clears the video. Score/transcript history may still be saved privately for progress.

By default for archiving, Speakwise does not keep practice audio or video as a long-term cloud library. Temporary files stay on your device for the session unless you choose to keep a copy. After you leave a session or refresh the page, that take cannot be replayed in Speakwise unless you exported it.

Cloud speech (on by default when free Groq is ready). Improve transcript with cloud speech in Settings → Voice & Audio defaults on when a free Groq path is ready (your Groq key in AI Features, or a platform Groq key). After you stop a Practice recording Speakwise may briefly send that audio through our Firebase Cloud Function to Groq Whisper to refine the transcript used for scoring. OpenRouter has no free speech tier and is only a paid fallback if Groq fails. That audio is processed for transcription only for the request and is not kept as a long-term recording archive on Speakwise servers. On-device speech recognition is still used for the live transcript while you speak. If cloud speech fails, your on-device transcript is kept. Turn the setting off and audio stays local for transcription.

Score and transcript history. After a scored practice attempt (and some Learn activities such as Role-Play), we may store score breakdowns, a short tag, and the transcript text in your private Firestore scoreHistory (and related activity history subcollections) under your account. This powers Progress, recommendations, and weak-spot tips. It is text history for progress — not a replayable audio/video archive.

AI feature usage. Signed-in users may use Speakwise AI features backed by platform keys held only on our Firebase Cloud Functions backend (never shown in the app) — Groq primary, then OpenRouter free models as fallback; no personal API key is required. Optionally, you may supply your own API keys (Groq, Cerebras, OpenRouter, Mistral, Gemini, OpenAI — "BYOK") as an override. Deeper AI scoring (a second refine pass; 2 API calls) is on by default and can be turned off in Settings → AI Features. When AI scoring or coaching runs, transcript text and prompts are sent to the provider that serves the request, subject to that provider's privacy policy. Audio and video are not uploaded for AI scoring prompts; cloud speech (above) uses free Groq Whisper when ready. Your own keys are cached locally and stored in your private Firestore account data; raw key values are omitted from "Download my data" exports.

App preferences. Notification settings, dark mode preference, Deeper AI scoring preference (on by default), recognition accent, cloud speech preference (on by default when free Groq is ready), practice-video preference (on by default locally), and consent choices.

Location (optional). If you use the Live Weather practice screen and grant permission, we read your device's approximate location once, to show local weather as a speaking-practice topic. This is not stored on our servers.

Biometric unlock. If you turn on "Unlock with Face ID / Fingerprint", your phone's own operating system handles the biometric check directly — Speakwise never receives or stores your fingerprint or face data.

App security signals (not product analytics). On supported platforms we use Firebase App Check (for example reCAPTCHA on web) so our backends can tell genuine app traffic from abuse. That is a security / abuse-prevention measure — not Speakwise product usage analytics.

2. Why we collect it

We use this information to create and secure your account; save and resume your learning progress; score your speaking practice from transcripts; personalize lessons; and send local reminders you've opted into on supported devices. Usage analytics are not currently collected — any in-app toggle is inactive / reserved for a future release.

3. Who we share it with

We do not sell your personal information. We share data only with the service providers that make the app work:

We may also disclose information if required by law, or to protect the rights, safety, or property of Speakwise or its users.

4. Data retention

We keep your account and learning data (including score/transcript history) for as long as your account exists. Practice audio and video are not stored as a long-term archive on Speakwise servers. Live Practice camera video is not archived. Score a recording video stays in session memory only (not Firebase Storage); extracted audio may be sent briefly for transcription. When cloud speech is enabled, audio is sent only for that transcription request. If you delete your account, your Firestore user document (including learning history), profile photo, and Firebase Authentication account are permanently removed.

5. Your choices and rights

6. Children's privacy

Speakwise is not directed at children under 13, and we do not knowingly collect personal information from children under 13. At sign-up we ask you to confirm you are 13 or older; this is self-attestation only and is not independent age verification. If you believe a child under 13 has created an account, please contact us so we can delete it.

7. Data security

Account credentials are handled by Firebase Authentication using industry-standard practices. Data in transit to Firebase and AI providers is encrypted (HTTPS/TLS). App Check helps reduce abuse of our backends. No method of transmission or storage is 100% secure.

8. Changes to this policy

We may update this policy as the app changes. We'll update the "Last updated" date above when we do. Continued use of the app after a change means you accept the updated policy.

9. Contact us

Questions about this policy or your data? Email us at wespeakwise@gmail.com.

See also our Terms of Service.