SpeakLab

Privacy Policy

Last updated: 2026-06-06. SpeakLab helps you practice spoken English with an AI conversation partner and personalized weakness diagnosis. This policy explains what we collect, why, and the controls you have.

What we collect

Microphone & speech recognition

SpeakLab needs microphone access so you can speak and get instant feedback. Your speech is converted to text using on-device speech recognition (Apple's SFSpeechRecognizer on iOS, the system speech recognizer on Android). The resulting text transcript — not raw continuous audio — is what powers the conversation. We do not stream your live microphone audio to our servers for general recognition.

How AI processing works

To generate a reply and corrections, your conversation text is sent through our private AI gateway (newapi.zweiteng.tk) to a large language model. Your text is used only to produce your response within the session and is not used to train third-party models. Request logs are retained for no more than 30 days for reliability and abuse prevention.

What we do NOT collect

Third-party services

Your rights (including GDPR)

Children

SpeakLab is not directed to children under 13. If you believe a child has used the app, contact settings@sanva.tk to request deletion.

Contact

Privacy questions: settings@sanva.tk. We respond within 30 days.

中文摘要

SpeakLab 用 Apple / Google 登录(仅取用户标识与可选邮箱);对话内容与语音转写文本、可选语音片段、弱项档案存于 Supabase(按用户隔离);语音识别在设备本地完成,仅将转写文本经 AI 网关处理;订阅经 RevenueCat / Apple;使用 Sentry/PostHog 做匿名诊断与分析(不含对话内容)。可在「设置 → 删除账号」一键彻底清除全部数据(符合 GDPR)。隐私问题联系 settings@sanva.tk。