Privacy policy

Developer & contact

MicMyDay is developed by (see the Terms Licenses Impressum for full postal details), which is responsible for the processing described in this policy within his control. Contact for privacy questions or requests.

MicMyDay does not require its own account, include advertising or analytics SDKs, or send your dictation to a MicMyDay-operated server. Optional providers receive data directly from your Mac when you use them. The account and service terms of your chosen provider also apply.

How dictation works

MicMyDay records microphone audio when you start dictation. If you enable voice activation, it listens while that mode is active and uses a short audio buffer to capture the start of speech. You can stop listening in the app.

Local Whisper and Parakeet

Downloaded models process recordings on your Mac. Model downloads connect to Hugging Face and its download infrastructure, which receive ordinary network information such as your IP address and the requested model. The download does not require sending your recordings or transcripts.

Apple Speech

With on-device recognition selected, MicMyDay requires local processing and reports an error if it is unavailable. If you allow server processing, Apple may receive your recording and vocabulary hints through its speech-recognition service.

Cloud or custom transcription

Your recording and the settings needed for transcription, such as model selection, language, and vocabulary hints, go to the selected endpoint. Google Gemini receives a version of the recording compressed on your Mac. A custom endpoint can run on your Mac, your network, or a remote server.

Optional AI rewriting

The transcript, rewrite instructions, and model selection go to the separately selected rewrite provider. Local transcription with cloud rewriting still sends text off your Mac. Rewriting does not require sending the original audio to the rewrite provider.

These requests serve the transcription or rewriting feature you choose. They can contain personal information spoken in a recording or included in vocabulary hints and rewrite instructions.

Providers & sharing

Before sending audio or text to a provider, MicMyDay shows the destination and content type and asks for approval. Approval is remembered for that destination and content type. You can cancel a request or reset approvals in Settings → Permissions. This also applies to custom and local-server destinations.

API keys or other configured credentials accompany provider requests. The provider receives connection information, including your IP address, and may associate requests and usage with your account. Gemini model discovery also contacts Google using your API key.

Provider retention, abuse monitoring, human review, and any use of content to improve models depend on the service, endpoint, account tier, region, and your settings. Review the linked terms before sending sensitive content. MicMyDay cannot promise zero retention or no training across all provider configurations.

For Gemini transcription, MicMyDay asks that interactions not be stored as retrievable conversations. This does not override Google's processing or retention terms. Deleting local files or history does not delete copies already held by a provider.

Providers may process data in countries other than your own. Their terms explain the locations and safeguards that apply to your account. Public internet endpoints require HTTPS. Plain HTTP is allowed for local servers and does not encrypt traffic, even on a home network.

Storage & retention on your Mac

Recordings

Audio uses temporary files in the app's sandbox. MicMyDay attempts to remove them after processing or cancellation. A crash or interrupted operation can leave temporary files for macOS to clean up. The app does not maintain an audio recording library.

Recent transcripts

If enabled, recent entries are saved in a local history file in the app’s sandbox and restored when MicMyDay reopens. You can configure how many entries to keep in Settings. Entries include spoken and delivered text, time, engine, rewrite profile, and destination app name. The history file has owner-only permissions and is excluded from backups. You can remove individual entries or clear all history. Turning off Keep recent transcripts stops new history entries and deletes the stored history file. The app may still hold entries and its most recent result in memory until you clear them or quit.

Clipboard and pasted text

Delivered text is copied to the macOS clipboard and, if allowed, pasted into your chosen app. Clipboard content can remain after MicMyDay quits and may be accessed or synced by macOS, clipboard managers, or other apps. Replace it separately if needed. The destination app controls text you paste into it.

Settings, prompts, and models

Preferences, sharing approvals, and custom rewrite prompts are stored locally. Downloaded models remain in app storage until removed. You can edit settings and delete models in MicMyDay.

Credentials

API keys, tokens, and custom request-header values are stored in the macOS Keychain. Clear credentials in provider settings before removing the app if you want to remove them; uninstalling an app does not necessarily delete Keychain entries. Revoke credentials with the provider to stop further use.

macOS permissions

• Microphone: record audio for dictation and listen while optional voice activation is active.
• Speech Recognition: use the Apple Speech engine.
• Accessibility / event posting: send the paste shortcut to your chosen app. Manual clipboard paste remains available without this access.
• Input Monitoring: detect modifier-only shortcuts. MicMyDay listens for modifier changes for this feature, rather than typed text.
• Local Network: connect to a provider you configure on your network.

Change these permissions in System Settings → Privacy & Security. Revoking access can disable the related feature. Sharing approvals in MicMyDay are separate from these macOS permissions.

Support, purchases, and website data

When you email support, the developer and the email services delivering the message receive your email address, your name if provided, the message, and any attachments. This information is used to answer your request, investigate the issue, and handle related follow-up. Send only what is needed and remove credentials and sensitive dictated content.

Support correspondence is retained for as long as needed to handle the request and related follow-up, or to meet applicable legal obligations. You can request deletion by email.

Direct purchases take place on Polar’s checkout. Polar processes checkout and payment information under Polar’s privacy policy. This website does not collect your payment-card details.

Cloud-provider billing is handled through your separate provider account.

This website uses Cloudflare Web Analytics for aggregate page-view and performance measurements. Cloudflare states that this service does not use cookies or browser storage, fingerprint visitors, or collect or use visitors’ personal data.

This website does not embed advertising trackers, third-party fonts, or a contact form. Cloudflare Web Analytics does not set cookies or use browser storage. The server delivering the website receives connection and request information, such as your IP address, requested page, time, and browser details, and may retain operational logs to deliver and secure the site. External links connect you to those sites only when you follow them.

Your choices & rights

You control the transcription engine, rewrite provider, recording mode, permissions, and sharing approvals. Use a downloaded local model with rewriting off for dictation without sending audio or text to a provider. To stop future sharing, stop using that provider or choose a local configuration. Reset approvals in Settings → Permissions to require approval again.

Where European data-protection law applies, we process support information to respond to your request and on our legitimate interest in providing and improving support. Website delivery, security, and aggregate website analytics rely on our legitimate interest in operating and improving the site. Legal obligations can require certain records to be retained. Optional provider sharing requires your approval; the provider explains its own legal bases in its terms.

Depending on the law that applies to you, you may request access, correction, deletion, restriction, or portability of personal data, object to processing based on legitimate interests, and withdraw consent where processing relies on it. Withdrawal affects future processing, not processing already lawfully carried out. You can also complain to your local data-protection authority.

MicMyDay is not directed to children. It is intended for users aged 13 and over, and the developer does not knowingly collect personal data from children under 13. If you believe a child has provided personal data, contact the address below and it will be deleted.

The developer does not sell personal information, does not share it for cross-context behavioural advertising, and does not use it for profiling or automated decision-making. MicMyDay contains no advertising or analytics SDKs.

Email to exercise rights concerning data the developer holds. We may need information to verify and locate your request. We do not have remote access to your local recordings, history, or provider accounts. For provider-held data, contact the provider; see steps for clearing local data and credentials.

Changes to this policy

This page will be updated when MicMyDay's data handling changes, with the revision date shown above. Review it when changing providers or enabling new processing features. For questions about a change, contact the developer.