Tinnitune — Tinnitus Synthesizer & Acoustic Therapy

🔗 Live Demo: tinnitune.naudycastellanos.com

Tinnitune is an interactive web-based tool designed for sound exploration, tinnitus frequency mapping, and custom masking sound therapy. Built natively using HTML5, CSS3, Vanilla JavaScript, and the Web Audio API, it delivers a seamless, zero-dependency audio experience with a premium, glassmorphic dark mode user interface.

---

🚀 Key Features

1. Tinnitus Tuner

• Selectable Waveform: Pure Tone (sine wave), narrowband noise (resembling a sharp whistle), and broadband noise (pink or white noise).
• Precise Frequency Control: Logarithmic coarse slider (20 Hz - 16 kHz) combined with fine tuning adjustments (±100 Hz).
• Tremolo Modulation: Pulsating zumbido simulation via a Low Frequency Oscillator (LFO) with adjustable rate and depth.
• Stereo Balance & Volume: Independent controls to shift the sound location or adjust the generator's volume.

2. Masking Mixer & Clinical Noise

• Therapeutic Notch Filter: When enabled, it dynamically cuts out a narrow band around your matched tinnitus frequency from all background soundscapes. This promotes neural retraining by giving fatigued auditory receptors a rest.
• Synthesized Soundscapes: Ocean waves (lowpass pink noise modulated by LFO), gentle rain (brown noise with high-pass droplets), and forest wind (slow LFO modulating a bandpass filter).
• Clinical Noise Generators: Playback of dynamically generated White, Pink, and Brown noise.

3. Residual Inhibition (RI)

• A circular visual timer (30s, 60s, 120s) that plays your matched tinnitus frequency at a similar volume level to test residual inhibition (the temporary reduction or elimination of tinnitus perception that occurs after auditory stimulation).

4. Wave Phase Experiment (Physical ANC)

• Demonstrates constructive (0° in phase) and destructive (180° out of phase) wave interference. Learn how sound waves collide in physical space and cancel each other out when using external stereo speakers.

5. Preset Manager (Saved Settings)

• Saves your current configurations (waveform type, frequencies, LFO rates, ambient volumes, and Notch Filter status) locally in the browser (localStorage) for quick one-click load or deletion.

6. Sleep Timer with Smooth Fade Out

• Set a sleep timer (15, 30, 45, or 60 minutes) to play therapy soundscapes as you sleep.
• Gentle Fade Out: During the last 60 seconds of the timer, the master volume declines exponentially to 0 to prevent waking the user up, resetting safely to your initial volume level afterward.

---

🛠️ Audio Architecture & Routing

Tinnitune leverages the power of the Web Audio API to process audio in real-time. The following diagram illustrates the routing path and connections of the audio nodes:

graph TD
    %% Synthesizer Input Nodes
    subgraph Synthesizer
        A[Tinnitus Generator<br>Oscillator or Filtered Noise] --> B[Modulation Gain<br>LFO Tremolo]
        B --> C[Generator Volume]
        C --> D[Stereo Panner]
    end

    %% Therapy & Masking Nodes
    subgraph Therapy & Masking
        E[Synthesized Soundscapes<br>Rain, Waves, Wind] --> H[Therapeutic Notch Filter<br>Cut at Tinnitus Freq]
        F[Clinical Noise<br>White, Pink, Brown] --> H
    end

    %% Phase Experiment Nodes
    subgraph Phase Experiment
        I[Phase Oscillator] --> J[Left Channel Gain]
        I --> K[Right Channel Gain<br>Inversion to -1.0 at 180°]
        J --> L[Stereo Merger]
        K --> L
    end

    %% Analyzer & Destination Connections
    D --> M[Spectrum Analyser<br>FFT Canvas]
    H --> M
    L --> M
    
    M --> N[Master Gain<br>General Volume]
    N --> O[Audio Destination<br>Speakers / Headphones]

    style H fill:#06c6e3,stroke:#fff,stroke-width:2px,color:#000
    style M fill:#a21cf3,stroke:#fff,stroke-width:2px,color:#fff

Loading

---

🖥️ Local Execution

Browsers restrict some Web Audio features and buffer loads when running directly from local files (file://). Running a lightweight local HTTP server is recommended:

Method 1: Using Node.js (npx)

1. Launch the server from your project folder:

   npx http-server

2. Open the URL shown in your console (usually http://localhost:8080).

Method 2: Using Python

1. Start the HTTP server from the command line:

   python -m http.server 8000

2. Open http://localhost:8000 in your web browser.

---

🌐 Deployment

This is a fully static web application and can be hosted on any modern static hosting provider.

Deploying to Cloudflare Pages

1. Connect Repository: Connect your GitHub repository to Cloudflare Pages via the Cloudflare Dashboard.
2. Build Configuration: Since this is a vanilla HTML/JS project, configure the build settings as follows:
   • Framework preset: None
   • Build command: Leave empty (no build step required)
   • Build output directory: / (root directory)
3. Deploy: Click Save and Deploy.
4. Custom Domain:
   • Go to your Page's project settings -> Custom domains -> Set up a custom domain.
   • Enter your desired subdomain (e.g., tinnitune.naudycastellanos.com).
   • If your domain naudycastellanos.com is managed by Cloudflare DNS, Cloudflare will automatically set up the CNAME record for you. If managed externally, add the provided CNAME record in your DNS registrar.

---

📄 Project Files

• index.html: Semantic page structure, SVG icons, visualizer canvas, and panels.
• style.css: Vanilla CSS styling featuring CSS variables, glassmorphic elements, and a responsive grid.
• app.js: Main JS file handling Web Audio nodes, canvas render loop (requestAnimationFrame), and DOM event listeners.
• CHANGELOG.md: Chronological log of versions and features added.
• Docs/: Directory for extended documentation. Contains future_development_ideas.md detailing clinical sound therapy expansion concepts (audiograms, symptom trackers, pulsed masking, etc.).

---

⚠️ Medical Disclaimer

This application is an educational sound exploration and therapeutic support tool. It is not a medical device, does not provide diagnostics, and does not replace the advice, diagnosis, or treatment of a health professional or audiologist. Use headphones at safe volumes to avoid further hearing damage.

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.