Privacy Policy

Non-binding translation. The is legally binding; in case of any discrepancy, the German version prevails.

1. Introduction

XSpeak is a privacy-first AI meeting assistant. We are committed to protecting your privacy by ensuring that all your data stays on your device (unless you choose to use the OpenAI-Compatible API). This Privacy Policy explains how XSpeak handles your information. It covers both the website and the app; some sections apply to one of them only.

2. Definitions

Text-based meeting data means the text information that XSpeak uses for AI analysis: your meeting transcripts, including the speaker names; the configured speech language and translation language; the configured meeting type (context); the configured prompt; and, where applicable, the results of previous AI analyses.

3. Local Data Handling

  • XSpeak does not collect personal user data through the app. The only transmission of your text-based meeting data is to the endpoint you configure, and only if you enable the OpenAI-Compatible API (disabled by default; see section 7).
  • The app does not access your contacts, identity, email, or any external accounts.
  • XSpeak does not use third-party analytics, tracking, or advertising SDKs.
  • During meetings, audio is temporarily recorded and held in memory, limited to the last 300 seconds.
  • Additionally, if the setting "Save Audio Recordings" is not disabled, selected audio fragments are saved locally to disk on your device. You can disable this at any time in Settings.
  • No audio is uploaded or shared externally by XSpeak.
  • If you choose to create a voice profile, XSpeak may store a local voice embedding (a numeric representation of your voice) only on your device. Additionally, when you label other speakers during a meeting, XSpeak learns their voices and stores their voice embeddings locally on your device as well.
  • All processing and storage happen entirely on your device unless you choose to use the OpenAI-Compatible API; no data is uploaded or shared externally unless you choose to use the OpenAI-Compatible API.
  • Meeting transcripts and AI analysis results are saved locally on your device. This data is never transmitted or shared externally unless processed by the OpenAI-Compatible API you configure.
  • You can delete all locally stored meeting data at any time from the app settings. XSpeak does not retain any of your data beyond your device.
  • If you choose to use the OpenAI-Compatible API, text-based meeting data (see section 2) may be sent to the endpoint you configure. In such cases, XSpeak does not guarantee the privacy or security of that data.

4. On-Device Processing

All core features of XSpeak run entirely on your device (unless you use the OpenAI-Compatible API):

  • Transcription of meetings is performed in real-time using on-device audio processing.
  • AI analysis (summarization and insights) is powered by an on-device language model.
  • If you create a voice profile, it is stored locally in the form of numeric voice embeddings (not audio recordings) and remains on the device as long as the app is installed; on macOS, the app's data container may remain on disk after the app is deleted. When you label other speakers, their voice embeddings are also learned and stored locally on your device.

All processing and analysis are performed locally by default. However, if you choose to use the OpenAI-Compatible API, text-based meeting data (see section 2) may be sent to the endpoint you configure for processing. In such cases, XSpeak does not guarantee the privacy or security of that data.

5. Recording Storage

XSpeak uses two forms of local storage for audio:

  • In-memory buffer: A rolling buffer of the last 300 seconds (5 minutes) kept in memory.
  • On-disk audio fragments: Selected audio fragments are saved locally to disk on your device. You can disable this in Settings at any time.
  • You can delete saved recordings and associated meeting data at any time from within the app.
  • No audio is uploaded, transmitted, or stored externally by XSpeak.

6. AI Analysis

AI analysis occurs when:

  • It's manually triggered by the user.
  • Real-time Insights are enabled.
  • When recording starts and there's enough context, the app may use the first statements to suggest a conversation name using AI.

7. OpenAI-Compatible API

The OpenAI-Compatible API is disabled by default. When you choose to use it for AI analysis, XSpeak connects to the endpoint you provide. In this case, only text-based meeting data (see section 2) is transmitted to and processed by the endpoint you configure. Audio and voice profiles (voice embeddings) are never sent. When the endpoint is enabled, this data is transmitted to it automatically for every AI analysis, including Real-time Insights and the suggested conversation name.

  • XSpeak sends the data to the endpoint address you configure. What happens to the data afterwards, including whether it is stored or forwarded to other services, depends on the endpoint and is outside XSpeak’s control.
  • XSpeak does not control or operate the endpoint you configure, and we have no access to the data sent to it.
  • The endpoint, or services to which it forwards data, may be located outside the European Economic Area (EEA). The privacy policy of the service that ultimately processes the data applies to you directly.
  • Therefore, XSpeak cannot guarantee the privacy, security, or compliance of any data processed outside your device.
  • By using the OpenAI-Compatible API, you acknowledge that data privacy depends on the endpoint you configure.

8. Your Control

Because all meeting data stays on your device (unless you use the OpenAI-Compatible API), you control it directly in the app:

  • View, manage, and delete your meetings, transcripts, audio recordings, and voice profiles at any time in the app.
  • Disable "Save Audio Recordings" at any time in Settings.
  • Deleting the app removes locally stored data from the app's storage on your device; iCloud, Time Machine, or other system backups may retain copies of this data. On macOS, the app's data container may remain on disk after the app is deleted.

9. Other Processing by XSpeak

App: XSpeak does not receive your data and is not the responsible party for the processing that takes place on your device.

In-app purchases: In-app purchases and subscriptions are handled by Apple Inc. Payment data is processed by Apple in accordance with Apple's privacy policy; Apple is solely responsible for this data.

Email: when you contact us by email, we process your email address, your name (if provided), and the content of your message (including any attached files) only to reply to you. The legal basis is Art. 6(1)(f) GDPR (legitimate interest). We store your email as long as it is needed to handle your request. After that, we delete it, unless we are legally required to keep it.

Website hosting: This website is static. We do not use forms, analytics, or cookies ourselves. It is delivered via Cloudflare. When you open a page, Cloudflare may process technical data such as your IP address and request data (for example, browser and time) in order to deliver and secure the site. We do not analyze or use this data ourselves. The legal basis is Art. 6(1)(f) GDPR (legitimate interest in the secure and stable delivery of the site). Cloudflare acts on our behalf under its data processing agreement. According to Cloudflare, it relies on its certification under the EU-U.S. Data Privacy Framework and on standard contractual clauses for international transfers; for details, see Cloudflare's GDPR information.

For the processing of your email correspondence and your visit to this website described above, the responsible party is the person named in the Imprint. For this data, you have the rights under Art. 15–21 GDPR. To exercise these rights, contact us using the contact details given in the Imprint. You also have the right to lodge a complaint directly with a supervisory authority at any time, in particular in the EU member state of your habitual residence, place of work, or the place of the alleged infringement.

10. Recording Other People

Meetings regularly involve other people. The audio, voices, and statements of other participants are personal data, for which you or your organization may be responsible, depending on how you use XSpeak. XSpeak cannot verify whether the participants have consented to being recorded. Obtaining their consent, where required, is your responsibility.

When you label other speakers, XSpeak also stores their voice embeddings locally on your device. Storing voice embeddings of other speakers may require their consent. Depending on the endpoint you configure for the OpenAI-Compatible API, you may need to inform the participants that text-based meeting data (see section 2) will be sent to the AI service you use.

11. Changes to This Privacy Policy

We may update this Privacy Policy from time to time. If we make significant changes, we will notify users within the app. The current version is always available at https://xspeak.app/privacy.

Last updated: November 14, 2026

12. Contact

If you have any questions about this Privacy Policy or how XSpeak handles your data, you can reach us using the contact details given in the Imprint.