Live Transcript & Translation
Transcribe, translate, and summarize on your device. After models are ready, core features work offline. Your conversation is never sent to developer servers.
wayaknow turns spoken words into live text, translates each completed sentence into a language you choose, and organizes the conversation into a clear summary.
Supported languages
Japanese, English, Korean, Simplified Chinese, Traditional Chinese, Spanish, French, German, Brazilian Portuguese, and Italian. Choose the spoken and target languages independently, even while listening.
Features
• Live transcription and sentence-by-sentence translation
• Domain dictionaries and custom word lists for each of the 10 supported languages (Full Access)
• On-device automatic detection between the selected transcription and translation languages by utterance (Full Access)
• Low Latency for keeping pace or High Quality for more natural wording
• Five summary formats: Standard, Meeting Notes, Lecture Notes, Research Presentation, and Guidance & Procedures
• Save transcripts, translations, and summaries as text
• An always-on-top mini view on Mac
• System Audio input (Full Access: Mac and iPhone/iPad with iOS/iPadOS 27 or later; use audio you have permission to process)
• View the latest iPhone transcript or translation on Apple Watch
Specialized vocabulary
Choose dictionaries for seven fields: Medicine & Life Sciences; Chemistry & Materials; Physics & Mathematics; Engineering & Industrial Technology; IT & Communications; Economics, Finance & Accounting; and Law & Public Administration. Engineering includes Mechanical & Manufacturing; Electrical, Electronics & Control; and Architecture & Civil Engineering. These settings work in all 10 supported languages. Register frequently used words separately for each language to help transcribe specialized terms.
Speech recognition, translation, and summaries use Apple-provided on-device frameworks and models. In Full Access automatic mode, a bundled whisper.cpp model identifies only the language of each utterance, then Apple SpeechTranscriber produces the transcript. Whisper transcription is used only as a fallback when Apple recognition is unavailable. Audio, transcripts, translations, and summaries are not sent to developer-operated servers. Apple models may need preparation on first use; once ready on a supported device, available core features can work without an internet connection. High Quality translation and summaries require an Apple Intelligence–compatible device and ready models.
In the free version, translation, summaries, and saving pause for one minute after every five minutes of microphone use. The one-time wayaknow Full Access purchase removes the wait and unlocks automatic language detection by utterance and System Audio input on supported devices. It is not a subscription. Full Access also includes terminology settings and custom words in all 10 supported languages.
Always verify important content against the original audio. Before processing third-party audio, confirm the required rights, permission, participant consent, service terms, copyright, privacy, and applicable law.