These are presets, not modes you are locked into — every part of what they set can be changed afterwards.
Every one of them is the same three decisions, answered differently: whose audio gets translated, whether the result is spoken out loud, and what ends up on your screen.
Online meetings, classes, talks, videos, streams — read a live translation of what you hear. Nothing is read aloud unless you switch on Translation I hear.
Who picks it: You are the listener, and the speaker is not going to meet you halfway.
sokuji translatesThey hear your translated voice through a virtual microphone. No install, no plugin, no captions to read on their side.
Who picks it: Anyone speaking in a language the room does not share, in a call they did not necessarily set up.
sokuji translatesTalks, streams, presentations — your audience reads translated subtitles; no audio is generated. You keep speaking in your own voice.
Who picks it: Presenters and streamers, where a second voice over your own would be worse than captions.
sokuji translatesThey hear your translated voice; you read theirs as subtitles and can hear it too. In the desktop app their translation is read aloud while Other's audio is one application, such as your meeting app; while it captures all system sound, you only read it.
Who picks it: A real back-and-forth where neither side shares a language.
sokuji translatessokuji translatesBilingual captions, meeting minutes — both sides as text, no synthetic voice.
Who picks it: Recorded meetings, note-taking, or any room where added audio would be unwelcome.
sokuji translatessokuji translatesTwo people at one computer, each speaking their own language; each wears one earbud and hears only the translation into their language. By default your earbud is the left channel and theirs the right.
Who picks it: Two people in the same room without a shared language — at a desk, a counter, a kitchen table.
sokuji translatessokuji translatesBoth read the same screen, or the other person opens the LAN caption page on their phone. The LAN caption page is in the desktop app only.
Who picks it: The same room, where a voice in the ear is unwelcome, or the other person would rather read on their own phone.
sokuji translatessokuji translates“Any app” is literal. On the way out Sokuji is an ordinary microphone, so anything that lets you choose one can carry your translated voice — a browser tab, a recorder, streaming software, a game. On the way in it works from whatever your machine is playing, system audio included, so a video, a stream or a recording behaves exactly like a call.
It turns on two properties. The first is whether the provider can produce a voice, and whether it can be kept quiet: one that returns only text cannot speak into the call, and one that always answers with audio cannot be held to subtitles. Every provider Sokuji offers today can do either, but the check stays for the next one. The second is face-to-face: two people at one microphone need a provider that can tell them apart, and only Soniox can, with your own key or through KizunaAI. Sokuji checks both when you choose the provider, so a scenario it cannot serve is closed off before the call rather than discovered during one.
available · not available with that provider
Understand what others say is the one that works everywhere: it starts as text, and the other side is read aloud only if you switch on Translation I hear. The last column is Soniox, with your own key or through KizunaAI.