Mersio

Immersive language learning through what interests you.

Bring a local video, a YouTube link, or a Bilibili link. Mersio turns it into a focused place to listen closely, understand every segment, and keep what you learn in context.

See how it works

Private preview · macOS (Apple silicon) · Invite required

A focused tool for learning from real-world video

  • On-device AI transcription
  • Meaningful segments
  • Contextual AI help
  • Editable transcript
  • macOS Apple silicon

Made for meaningful input

Stay with the language, not the tools.

Choose the material. Mersio handles the mechanics around it, while you decide how deeply to listen and explore.

Your interests, not a content catalog

Start with the video you actually want to understand.

A favorite interview, a technical talk, a film clip, or a travel story can become learning material. Import a local video or paste a supported link—Mersio does not choose the lesson for you.

  • Local video
  • YouTube
  • Bilibili

Local Whisper transcription

Your audio becomes a transcript on your Mac.

Mersio runs Whisper locally, then organizes speech into meaningful segments: small enough to replay, but long enough to keep a complete idea intact.

Audio stays on this Mac while Whisper transcribes it.

Listen your way

Stay with one segment until it clicks.

Replay the active segment, slow it down, or hide its words for a blind listen. Playback and transcript stay together, so repetition never turns into timeline hunting.

  • Segment loop
  • 0.75× Variable speed
  • Blind listening

Understanding on demand

Ask for help without leaving the moment.

Select a word, phrase, or passage for an AI explanation that includes the surrounding transcript. Turn on Simplified Chinese translation when you need a second view, then return to listening.

Explain and translation send the relevant transcript text—not your media—to Mersio's cloud AI.

A transcript you control

If Whisper gets it wrong, make the segment yours.

Edit the words directly. Split or merge segments when an idea was grouped poorly, and drag timing boundaries against the waveform when the audio starts or ends in the wrong place.

FAQ

Questions, answered.

Still unsure about something? The changelog tracks what shipped in every build.

Which Macs can run Mersio?

Mersio currently ships for macOS on Apple silicon. Intel Macs and other platforms are not supported yet.

Why do I need an invite code to download?

Mersio is in private preview. The invite code only unlocks the download here; you activate the app itself with the same code on first launch.

Does transcription happen in the cloud?

No. Mersio downloads Whisper during setup and transcribes audio on your Mac. AI Explain and translation are separate, on-demand cloud features that send relevant transcript text—not your media—to Mersio’s backend.

What can AI help me understand?

Select a word, phrase, or passage to get an explanation based on its surrounding transcript context. You can also translate individual segments or the full transcript into Simplified Chinese.

Where do my media, transcripts, and notes live?

Mersio keeps imported media in its managed media library on your Mac. Transcripts, segment timing, and notes are stored in a local database, with each note linked to its source segment.

Which sources can I import?

Local video files, YouTube links, and Bilibili links.

What if the transcript is wrong?

Every segment remains under your control. Edit its words, split or merge it with another segment, or adjust its start and end timing against the waveform.

Learn deeply from what already interests you.

Bring the next video you want to understand. Mersio gives you one quiet workspace to listen, explore, and make it yours.