ekto?

KINDAYou can build the valuable core, but the established product has meaningful operational complexity. Pick one narrow promise for v1.

BUILD FIRST

One specific customer type and one primary use case
A vertical slice that completes the main job without manual workarounds
Visible limits, review steps, and a fallback when an external dependency fails

LEAVE FOR LATER

Broad third-party integrations
High-scale collaboration, syncing, and permissions
Automation that can cause harm without human review

RESEARCH NOTES

What this tool actually does.

Checked 2026-09-06

Ekto (also marketed as 'Ekto - AI Live Interpreter' / 'Live Translator: ekto Captions') is an iOS voice-translation app by independent developer Anthony Ho. It converts spoken speech into real-time captions in the user's native language across 37 languages. Two modes: Live Transcribe & Translate (lectures/conferences, claims to work up to 10 meters away) and Conversation mode (flip view with automatic language switching for two-way dialogue). Translations appear every 3-6 seconds; transcript/translation history can be saved and reviewed; long talks can be auto-summarized. Freemium: free tier plus PRO plans ($59.99/yr Lite, $139/yr PRO, $8.99/wk) with daily limits (60-90 min). Requires iOS 17/macOS 14, internet connection, and microphone access; developer states no data collected. No official product website found; terms link to verby.co.

SOURCES

RESEARCHED BUILD PROMPT

Give this to your coding agent.

Build a web-based replacement for the iOS app 'Ekto' (ekto Captions / AI Live Interpreter), a real-time voice translator that shows live translated captions for lectures and two-way conversation. Core workflow: user presses one button, speaks in any of ~37 supported languages; the app displays near-real-time speech recognition and translation as incremental captions refreshing every 3-6 seconds; a second Conversation mode renders separate color-coded caption panes per speaker with automatic language detection/switching and a flip view for face-to-face use. User-facing requirements: (1) one-button start/stop; (2) Listen mode targeting distant speakers (up to 10 m) via microphone gain/far-field processing; (3) Conversation mode with speaker diarization using voice-activity energy thresholds; (4) live caption display with source text, translated text, timestamp, and wrap; (5) transcript save/export plus auto-summarization of long talks; (6) settings for source/target language auto-detect and UI mirror toggle for the flip view. Technical requirements: use MediaRecorder or Web Speech API for capture, streaming ASR (Deepgram streaming or Whisper API) for transcription, then LLM or Translate API for translation; chunk audio so captions appear every 3-6 s without waiting for silence; implement a session state machine (idle → listening → processing → caption appended); persist transcripts in IndexedDB or JSONL. Implementation approach: React + Vite front-end, Web Audio for gain/diarization, thin server or client-side API-key handling. Do not copy Ekto's brand, logo, screenshots, or UI layout; invent a clean caption-focused interface. Out of scope: native iOS app, offline translation, simultaneous interpreting of overlapping speech, multi-user accounts, full mobile SDK. Deliver a README covering setup, API-key environment variables, supported languages, latency behavior, and known limitations (distance accuracy, network dependency, non-overlapping speech only).

Visit the official product page