TalkScript vs WhoSpoke

Meeting notes that show who said what, finished entirely on your device. TalkScript turns meeting, interview, and lecture recordings into speaker-separated transcripts. Speech recognition and speaker diarization both run on-device, so your recordings are never sent to any server. Key Features - Speaker separation: automatically tells who said what (S1, S2 ...). Each speaker is color-coded; set the number of speakers, assign names, and merge incorrectly split speakers. - Subtitle mode: when you do not need speakers, transcribe as timestamped subtitles with text only (lectures, memos, single-speaker recordings). - Speaking-time stats: see who spoke how much, with each speaker's talk time, share, and turn count at a glance. - On-device speech recognition: Whisper-based multilingual models recognize about 99 languages including English and Korean. The right model is selected automatically for your device. - Record and import: record in the app or import m4a, wav, and mp3 files (up to 3 hours). - Synced playback: tap a sentence to play that moment, with the current sentence highlighted as it plays. Includes 0.5x-2x playback speed and 15-second skip. - Edit and search: fix text, reassign a sentence to another speaker, search inside transcripts, and adjust the body text size. - Export: share as plain text (TXT), Markdown (MD), subtitles (SRT), or a formatted PDF, or copy a single segment. - Organize: pin frequently used recordings as favorites for quick access. - Reliable long recordings: progress is saved while processing, and resumes from the last checkpoint if you leave the app. Thermal and storage conditions are managed automatically. Privacy First - Audio and transcripts are never transmitted off your device. - After a one-time model download, transcription works without internet. - Use it in secure facilities, on flights, or anywhere connectivity is limited. TalkScript Pro (Optional Subscription) All core features are free. For heavier users, an optional monthly subscription adds: - No ads - No per-transcription length limit (free up to 20 minutes, Pro up to 3 hours) Made For - Professionals who would rather not spend hours writing meeting minutes - Journalists, researchers, and creators who re-listen to interviews to take notes - Anyone whose recordings are too sensitive for cloud transcription services Platforms: iPhone, iPad, Mac Recognition languages: about 99, including English, Korean, Japanese, and Chinese Privacy Policy: https://cksw.pe.kr/talkscript/privacy.html Terms of Use: http://www.apple.com/legal/itunes/appstore/dev/stdeula

WhoSpoke turns your recordings into speaker-separated transcripts — entirely on your iPhone. No servers, no account, no data collection. Perfect for meetings, interviews, and lectures where knowing who said what matters. WHY WHOSPOKE Most transcription apps upload your conversations to the cloud. WhoSpoke never does. Recording, transcription, speaker separation, and even translation all run on device, so sensitive conversations stay exactly where they belong — with you. KEY FEATURES • Speaker separation: An on-device model tells voices apart and organizes the transcript by speaker • Live captions: watch the transcript appear in real time while you record • On-device transcription: supports 10 languages, powered by Apple's speech engine • Name suggestions: Apple Intelligence suggests speaker names from conversation clues, on device — you confirm • Tap to replay: tap any sentence to jump straight to that moment in the audio • Translation: translate transcripts into 20 languages, fully offline • Import audio: bring in existing recordings and analyze them the same way • Edit everything: fix transcripts, rename speakers, merge duplicates • Export: share as text or Markdown, or export the original audio • Smart titles: Apple Intelligence names your recordings based on what was discussed, on device BUILT FOR • Meetings — know who committed to what • Interviews — separate questions from answers automatically • Lectures & seminars — replay exactly what the professor said • Podcasts & research — turn multi-person audio into structured text PRIVACY FIRST WhoSpoke collects no data. There are no analytics, no ads, no tracking, and no third-party AI services — all processing runs on device using Apple's speech, translation, and Apple Intelligence frameworks. The app's only network request is a one-time download of the speaker-separation model files to your iPhone; nothing is ever uploaded. Try it now: load the bundled sample meeting from the + menu and see speaker separation in action within minutes. Terms of Use: https://www.apple.com/legal/internet-services/itunes/dev/stdeula/ Privacy Policy: https://finepocket.app/WhoSpoke/privacy