Close your eyes.
Visualize each word.
8s ·
20 min
Close your eyes.
Visualize each word.
Downloading voice…
The Cognitive Shuffle — also called Serial Diverse Imagining (SDI) — is a technique developed by cognitive scientist Dr. Luc Beaudoin at Simon Fraser University. It works by occupying your mind with random, unrelated mental images, which mimics the scattered thinking pattern your brain naturally produces as it falls asleep.
When you lie in bed with racing thoughts, your brain stays in “problem-solving mode.” By visualizing random, concrete words — like “umbrella,” then “forest,” then “bicycle” — you disrupt that pattern and signal to your brain that it’s safe to let go. The randomness is key: it prevents your mind from constructing a narrative that could keep you awake.
Research presented at the SLEEP 2016 conference found that participants using this technique fell asleep significantly faster than control groups.
Tap start, close your eyes, and listen. When you hear a word, picture it in your mind — the more vivid, the better. Don’t try to connect the words. Just let each new image replace the last. Most people fall asleep within 15–20 minutes.
Voice Technology
Speech synthesis by Piper ·
ONNX Runtime
Voice Data
LibriTTS corpus (CC BY 4.0)
Zen, H. et al. “LibriTTS: A Corpus Derived from LibriVox.” Interspeech, 2019.
Word Frequency
SUBTLEX-US
Brysbaert, M. & New, B. (2009). Behavior Research Methods, 41(4), 977-990.
Built with care and no ads, forever.
v1.1 · wordtosleep.com
By using Word to Sleep you agree to these terms. If you do not agree, please stop using the app.
Not medical advice. Word to Sleep is a relaxation aid, not a medical device or treatment. It is not intended to diagnose, treat, cure, or prevent any condition, including insomnia or sleep disorders. If you have a medical concern, consult a qualified health professional.
Provided “as is.” The app is provided without warranty of any kind, express or implied, including but not limited to warranties of merchantability, fitness for a particular purpose, or non-infringement.
Limitation of liability. To the fullest extent permitted by law, the authors and contributors shall not be liable for any direct, indirect, incidental, special, or consequential damages arising from your use of, or inability to use, the app.
No guarantee of availability. The app, voice models, and third-party services it depends on (CDN, Hugging Face) may become unavailable at any time without notice.
Intellectual property. The source code is released under the Apache License 2.0. The name “Word to Sleep” and associated branding remain the property of their owner.
These terms may be updated from time to time. Continued use of the app after changes constitutes acceptance. Last updated February 2026.
Word to Sleep does not collect, store, or transmit any personal data. There are no accounts, no analytics, no tracking, and no ads.
All data stays on your device. Settings are saved in local storage; the voice model is cached in IndexedDB for offline use.
Network requests are made only to load the app, download the voice model from Hugging Face, and load ONNX Runtime from a CDN — all on first use only. After that, the app works entirely offline. No data is ever uploaded from your device.
Word to Sleep is open-source software released under the Apache License 2.0. You are free to use, modify, and distribute this software, provided you include the original license and copyright notice.
Copyright 2025 Word to Sleep · Full license text
Speech synthesis engine by Michael Hansen. Word to Sleep uses pre-trained ONNX model files only.
Neural network inference engine by Microsoft.
Multi-speaker speech corpus derived from LibriVox audiobooks.
Zen, H. et al. “LibriTTS: A Corpus Derived from LibriVox.”
Interspeech, 2019.
Reference implementation for browser-based Piper TTS.