👆 Hand Control

Control audio with hand gestures

Loading MediaPipe Hand Detection... Please wait.

📱 Touch Controls

Touch, tap, or scroll here

Control Mode:

🔄 Circle 📊 Swipe Freq. ⚡ Swipe Speed

Circle/Tap

🔄 Circles = Rewind/Forward • 👆 Taps = Forward

Touch Parameters

▼

🎚️ Shared Settings

Gesture Amplitude: Overall sensitivity

Touch Alpha (α): 0.010 Smoothing speed (higher = faster response)

Decay Speed (α): Return rate

Speed Decay Countdown: 2.00s Time until decay starts

👤 Facial Expressions Recognition

▼

Enable Face Control Confusion adjusts decay speed

🎛️ Control Mode

▼

Distance Mode Rotation controls speed Speed Mode Fingertip velocity controls speed Touch Mode Touch/mouse/scroll controls

🤚 Control Instructions

▼

👆 Point index finger up to enable gesture control
↻ Rotate clockwiseto go forward
↺ Rotate counter-clockwiseto go backwards
🔄 Two full rotations to start playback from a stopped state
✋ Open palm to pause playback
🔊 Raise your other hand and point index finger up/down to control volume - finger up = louder, finger down = quieter

🎵 Audio Controls

📁 Load Audio

0:00 0:00

Input Speed: 1.00x

◄ Backward (-4x) Forward (4x) ►

Playback Speed: 1.00x

Smoothing Parameters

▼

Active Alpha: 0.025

Hand detected - Responsive mode

Alpha (Finger Detected): 0.01 Responsive tracking

Alpha (No Finger): 0.001 Slow drift

Drift Target Speed: 0.70x Target when idle (positive only)

Gesture Amplitude: 1.00 Gesture sensitivity (scales input)

📝 Transcription & Smart Scrubbing

▼

Smart Scrubbing Settings

Semantic Threshold: 80th percentile

Higher = fewer, more important words

Activation Speed: 2.0x

Speed at which smart scrub activates

Word Interval: 0ms

Gap between words (0 = auto based on speed)

Word Playback Speed: 1.0x

Speed for individual word playback

No keyword playing

Navigation Parameters

▼

Segment Duration (s) Chunk duration

Segment Step (s) Distance between chunks

Segment Interval (ms) Time between chunks

Fade Duration (s) Crossfade time

Effective Values (at current speed):

Step: 1.5s Interval: 1200ms

📊 How it works

▼

• Input Speed: Controlled by hand gestures (raw input)

• Playback Speed: Smoothed speed (what you hear)

• Formula: s = s - α(s - A·g) where A is amplitude

• Alpha (Finger): Fast response when hand detected

• Alpha (No Finger): Slow drift towards target

• Amplitude: Scales gesture sensitivity (higher = more sensitive)

• Forward (positive): Normal playback with tempo control

• Backward (negative): Chunk-based navigation

• Speed adjustments preserve pitch in both directions

👆 Hand Control

📱 Touch Controls

Touch Parameters

👤 Facial Expressions Recognition

🎛️ Control Mode

🤚 Control Instructions

🎵 Audio Controls

📁 Load Audio

Smoothing Parameters

📝 Transcription & Smart Scrubbing

Smart Scrubbing Settings

Transcript:

Navigation Parameters

📊 How it works

⚙️ Parameters

📁 Load Audio

✓ Audio Loaded

📂 Upload from File

🎬 Load from YouTube

📝 Default Audio