Caption.IM

Caption.IM is a privacy-first AI captioning assistant designed exclusively for macOS, transforming how you interact with audio on your computer. It converts any audio source into real-time subtitles, instant translations, structured meeting notes, and searchable recordings, all processed locally on your device for maximum privacy and speed. Unlike browser extensions or meeting bots that require integration with specific platforms, Caption.IM captures system audio directly, making it compatible with virtually any application you use: Zoom, Google Meet, Microsoft Teams, YouTube, online courses, podcasts, livestreams, webinars, and pre-recorded videos. The product is built with local AI and Local LLMs at its core, ensuring your conversations never leave your Mac while delivering ultra-fast speech recognition optimized for Apple Silicon (M1, M2, M3, and later). Whether you are a remote worker struggling to keep up with fast-paced meetings, a student trying to follow lectures in a second language, a content creator needing accurate transcripts, or someone with hearing accessibility needs, Caption.IM provides an elegant, frictionless solution. The floating subtitle window integrates seamlessly with macOS, offering a transparent overlay that stays out of your way while keeping you informed. With continuous improvements through iterative updates, Caption.IM is constantly refining its audio pipeline, transcription accuracy, and user experience to deliver an ever-improving tool for turning any conversation into searchable, translatable knowledge instantly.

Real-Time Transcription

Caption.IM generates live captions for any audio playing on your Mac, whether from video calls, podcasts, recorded videos, or online courses. The transcription engine runs locally on your device using optimized speech recognition models, delivering minimal latency and high accuracy. As you speak or listen, words appear in real time on a floating subtitle window, allowing you to follow conversations without missing a beat. This feature is particularly valuable for fast-paced meetings or lectures where taking notes manually is impractical. The system continuously improves through updates, with recent enhancements to the audio pipeline including source-stage 16 kHz mono Float32 conversion for even better accuracy.

Instant Translation

Break down language barriers with real-time translated subtitles that work across multiple languages. Caption.IM can translate spoken content as it happens, displaying the translated text alongside or instead of the original captions. This makes it an indispensable tool for multilingual teams, international meetings, or consuming foreign language content like podcasts and online courses. The translation engine operates locally, ensuring sensitive conversations remain private while delivering instant results. Whether you are collaborating with colleagues across borders or learning from global content creators, Caption.IM helps you understand and engage with information in any language.

Floating Subtitle Window

An elegant, transparent overlay that works seamlessly with macOS, the floating subtitle window is designed to be both functional and unobtrusive. It stays on top of other applications without blocking your view, allowing you to read captions while continuing to work, watch videos, or participate in calls. The window can be repositioned anywhere on your screen and resized to suit your preferences. Its clean, minimal design integrates naturally with the macOS aesthetic, making it feel like a native part of the operating system. This feature ensures you never have to switch between windows or lose context when following audio content.

AI Meeting Summaries

After any conversation, Caption.IM automatically generates structured summaries, key points, action items, and even mind maps from your audio recordings. This transforms long discussions into concise, actionable insights that you can review, share, or save for later. The local AI processes the transcription data to identify important topics, decisions, and follow-up tasks, saving you hours of manual note-taking. Whether it is a one-on-one meeting, a team stand-up, or a lengthy webinar, you get a clear, organized summary that captures the essence of the discussion. This feature is continuously refined to improve summary accuracy and structure with each update.

Remote Meetings and Video Calls

For professionals working remotely, Caption.IM provides real-time captions for Zoom, Google Meet, Microsoft Teams, and other video conferencing tools. You can follow conversations more easily, especially in noisy environments or when multiple people are speaking. The AI meeting summaries ensure you never miss action items or key decisions, even if you join late or need to step away. This use case is ideal for project managers, team leads, and anyone who participates in frequent virtual meetings and needs to stay organized.

Online Learning and Education

Students and lifelong learners can use Caption.IM to generate live subtitles for online courses, lectures, and educational videos. The instant translation feature helps non-native speakers understand content in other languages, while the recording and summary capabilities make it easy to review material later. Whether you are taking a course on YouTube, watching a recorded lecture, or participating in a live webinar, Caption.IM turns audio into text you can search, highlight, and revisit, enhancing comprehension and retention.

Multilingual Team Collaboration

In global organizations where team members speak different languages, Caption.IM bridges communication gaps with real-time translation. During international meetings, participants can read translated subtitles in their preferred language, reducing misunderstandings and ensuring everyone stays aligned. The local processing ensures that sensitive business conversations remain private and secure. This use case is essential for HR teams, global project managers, and executives working across borders.

Content Creation and Research

Content creators, journalists, and researchers can leverage Caption.IM to transcribe interviews, podcasts, and recorded discussions accurately. The floating subtitle window allows you to monitor audio in real time, while the recording and summary features help you organize findings into actionable notes. For researchers conducting interviews or analyzing focus groups, the ability to generate structured summaries and mind maps saves significant time compared to manual transcription. This use case supports anyone who needs to capture and process spoken information efficiently.

Does Caption.IM work with any application on my Mac?

Yes, Caption.IM captures system audio directly, so it works with virtually any application that produces sound on your Mac. This includes video conferencing tools like Zoom, Google Meet, and Microsoft Teams, as well as media players, web browsers, podcast apps, and online course platforms. Unlike browser extensions that are limited to specific websites or meeting bots that require integration, Caption.IM operates at the system level, making it universally compatible. The only requirement is that your Mac runs macOS 15.6 or later and uses Apple Silicon (M1, M2, M3, or later) for optimal performance.

Is my audio data private when using Caption.IM?

Absolutely. Caption.IM is built with a privacy-first architecture that processes all speech recognition and translation locally on your device. Your audio data never leaves your Mac, meaning no third-party servers are involved, no recordings are uploaded to the cloud, and no bots join your meetings. This local processing ensures that sensitive conversations, whether personal or professional, remain completely confidential. The developer, INNAIO France, does not collect any data from the app, as confirmed in the privacy policy. You maintain full control over your information at all times.

Can I use Caption.IM for languages other than English?

Yes, Caption.IM supports real-time translation for multiple languages, allowing you to understand content in different languages as it is spoken. The instant translation feature displays translated subtitles alongside or instead of the original captions, making it ideal for multilingual teams, international meetings, and foreign language content. While the app interface is currently in English, the underlying AI models can process and translate various languages. The local processing ensures that translations are fast and private, with continuous improvements to accuracy through regular updates.

How do I get started with Caption.IM?

Getting started is simple. Download Caption.IM from the Mac App Store, install it on your Mac (requires macOS 15.6 or later and Apple Silicon), and open the application. The floating subtitle window will appear, ready to capture system audio from any app you use. There is no complicated setup, browser dependency, or need to invite bots to your meetings. Just launch the app, start playing audio from your desired source, and captions will appear in real time. The app is free with in-app purchases for additional features, and subscriptions automatically renew unless canceled at least 24 hours before the end of the current billing period.