What Technology Does Say After Me Use?

Say After Me is built on three core technologies: OpenAI text-to-speech voice synthesis for delivering affirmations in natural, warm human-sounding voices; Apple's on-device Speech framework for recognizing when users speak affirmations aloud; and an intelligent tracking system that monitors consistency, tracks streaks, and supports long-term habit formation. This technology stack creates a unique interactive experience where AI guides you through affirmations and speech recognition confirms your active participation — transforming affirmation practice from passive consumption into engaged, spoken commitment.

OpenAI Voice Synthesis

Say After Me uses OpenAI's text-to-speech models to generate the guide voices that deliver each affirmation. Unlike traditional text-to-speech that sounds robotic and disconnects the listener, OpenAI's neural voices capture natural intonation, warmth, and emotional nuance that make guided sessions feel genuinely supportive. The app offers six distinct voices, each with its own character: Alloy (neutral and balanced, included in the free tier), Echo (warm and engaging), Fable (expressive and storytelling), Onyx (deep and authoritative), Nova (bright and friendly), and Shimmer (soft and soothing). Premium users can switch between all six and adjust speech rate and pitch to match their preference. Earlier versions of the app used ElevenLabs for voice synthesis, but Say After Me migrated to OpenAI's text-to-speech platform, which delivers the vocal naturalness the practice requires. Research on voice processing shows that this naturalness directly impacts emotional engagement, ensuring that every affirmation lands with the emotional weight it deserves.

Apple Speech Recognition

The second core technology is Apple's Speech framework, which provides on-device speech recognition that processes your voice locally on your iPhone without sending audio to external servers. When Say After Me prompts you to repeat an affirmation, the speech recognition system listens, transcribes your speech, and confirms that you actively said the words, using a fuzzy-matching threshold of roughly 70% similarity so natural variations in phrasing do not block your progress. This creates a true repeat-after-me interaction — you hear the affirmation in a natural AI voice, then you say it aloud, and the app confirms your participation before moving to the next one. The on-device processing ensures complete privacy; your spoken affirmations never leave your phone.

The Repeat-After-Me Experience

The combination of voice synthesis and speech recognition creates something unique in the affirmation app space. Most affirmation apps either display text for you to read or play audio for you to listen to — both are passive experiences. Say After Me's repeat-after-me format requires active vocal participation, which cognitive science research shows is significantly more effective. The production effect — the memory and emotional advantage of speaking aloud — increases the impact of affirmations by 20-30% compared to silent reading. By requiring you to physically speak each affirmation, Say After Me ensures you are fully engaged rather than going through the motions. Each session also receives a conviction score from 0 to 10, computed from your volume (40%), pace (30%), and hesitation (30%), giving you concrete feedback on how convincingly you delivered each statement.

Habit Tracking and Streak System

Beyond the core voice technologies, Say After Me includes a habit tracking system designed around behavioral science principles. The app tracks daily completions, maintains streak counts, and provides visual progress indicators that leverage loss aversion psychology to encourage consistency. The tracking is automatic — every completed session is logged without requiring any additional action from the user. This removes the friction of separate habit-tracking apps and ensures that the accountability system is seamlessly integrated into the practice itself.

Built for iOS

Say After Me is built natively for iOS using Swift and SwiftUI, taking full advantage of Apple's hardware and software ecosystem. Native development ensures smooth performance, efficient battery usage, and deep integration with iOS features like notifications, widgets, and the Neural Engine that powers on-device speech recognition. The app is designed to be fast, responsive, and frictionless — because even small delays or technical frustrations can become excuses to skip a day of practice.