Lull: Tell It What's Bothering You, and AI Writes You a Meditation on the Spot

A Meditation App That Doesn't Play Recordings

Most meditation apps follow the same logic: browse a library of preset themes and follow along with a pre-recorded audio track. Lull takes the opposite approach. Instead of playing any pre-recorded content, it asks you to spend a minute describing what's actually troubling you right now — then uses AI to instantly generate a meditation tailored to that specific situation, read aloud in one of 11 available voices.

The difference seems subtle, but it cuts to the heart of a long-standing pain point in meditation products: generic recordings rarely match a person's emotional state in the moment. When you're anxious about an upcoming interview, or lying awake at night replaying a conversation, a vague "relax your mind and body" guide often feels like it's missing the point entirely. Lull is trying to close that gap with generative AI.

How Biometric Data Shapes the Meditation Experience

Another standout design choice in Lull is how it weaves biometric signals from wearable devices into a complete feedback loop.

Oura Recovery Data Adjusts the Tone

According to Lull's Product Hunt page, the app reads recovery data from your Oura Ring to adjust the tone of each meditation. When your body is in a low-recovery state, the generated content and pacing may lean toward the gentle and soothing; in higher-recovery states, a different guidance style might be used. It's an attempt to translate "how your body feels today" directly into content-generation parameters.

Oura's Recovery Score is a 0–100 composite metric calculated from the previous night's heart rate variability (HRV), resting heart rate, body temperature deviation, and sleep quality — used to estimate how well your body can handle additional stress that day. HRV is the most critical input: when the autonomic nervous system is parasympathetically dominant, HRV is higher, signaling a state of recovery; when sympathetically dominant, HRV drops, often indicating fatigue or stress. Lull maps this pre-calculated "today's status" score directly into the tone parameter for content generation — bypassing the complexity of interpreting raw physiological signals and instead feeding Oura's already-processed conclusion as part of the prompt to the underlying model. It's a pragmatic engineering integration.

Apple Watch Records Whether You Actually Calmed Down

Perhaps even more notable is Lull's honest approach to measuring outcomes. The app uses Apple Watch to record heart rate changes during meditation, then shows you afterward whether your body actually settled down — or whether "nothing" happened. That phrase from the product copy — "the real settle after, or nothing" — reveals a rare restraint: it doesn't assume every session is effective, but instead puts objective data in front of you and lets you decide.

This approach of using physiological feedback to validate subjective experience nudges meditation away from something that "feels mystical" and toward something measurable and verifiable.

The reason heart rate changes can serve as a proxy metric for meditation effectiveness lies in the autonomic nervous system's regulatory mechanisms: effective relaxation practices typically activate the parasympathetic nervous system, leading to a drop in heart rate and a rise in HRV. Apple Watch continuously records heart rate via a photoplethysmography (PPG) sensor, and the HealthKit interface in the Health app allows third-party apps to read this time-series data. Lull uses this data stream to compare baselines before and after meditation, providing an objective "settled or not" assessment. Worth noting: variables beyond meditation itself — posture, breathing rhythm, ambient noise — can all influence heart rate during a session, so this metric is best treated as a long-term trend reference rather than a definitive verdict on any single session.

Product Positioning and Market Signals

Lull was built by Evan Lane and launched on Product Hunt, earning 87 upvotes and 2 comments, ranking 5th for the day, and listed under three categories: Apple Watch, Health & Fitness, and Meditation.

Those categories make its positioning clear: it sits squarely at the intersection of the Apple ecosystem and health hardware, targeting users who already own an Apple Watch or Oura Ring and have some affinity for the quantified self movement. This is both a strength and a barrier — hardware dependency narrows the potential user base, but it also enables a level of personalization and feedback that pure-software meditation apps simply can't replicate.

The Opportunity and Risk of AI-Generated Content

From a technical perspective, Lull is a textbook example of generative AI landing in a vertical use case: combining a large language model's text generation capabilities with voice synthesis (11 voices) and biometric data input to produce personalized output tuned to a specific moment.

The opportunity lies in infinite personalization across time and state — the same person gets a different meditation depending on when and how they're feeling, something a recording library can never deliver. The risks are equally real: will AI-generated meditation scripts maintain consistent quality? Could the guidance come out sounding stilted or tonally off? In emotionally sensitive contexts, the reliability of generated content matters enormously. The source material doesn't include any in-depth user experience reviews, so these questions will need to be answered through real-world use.

Text-to-speech (TTS) technology is an easy-to-underestimate link in Lull's experience chain. Meditation guidance places far higher demands on voice quality than standard audio playback: pacing, pause timing, and tonal variation all directly affect how well a listener relaxes, and early TTS systems often struggled in these dimensions. Next-generation neural TTS services like ElevenLabs can now produce speech with nuanced emotional inflection, with adjustable parameters like "stability" and "similarity boost" for style customization — making it genuinely possible to craft a voice suited to a meditation context. Lull's 11 voice options suggest meaningful investment in vocal variety, but dynamically matching the rhythm of AI-generated text to the voice — such as automatically lengthening pauses at emotional low points — remains a detail that still needs refinement.

Takeaway

Lull's core innovation isn't simply "using AI to write meditations." It's the way it connects what you're thinking right now (voice input), how your body is feeling right now (Oura recovery data), and whether the meditation actually worked (Apple Watch heart rate feedback) into a complete, personalized loop. For users who are tired of one-size-fits-all meditation recordings and are willing to invest in a hardware ecosystem, Lull represents a genuinely new direction worth watching.