Getting Started with Pose, Gaze and Speech Detection

Modified on Fri, 12 Sep at 3:29 PM

Pose Detection: How to Position the Child


Purpose: Detects body movements and gestures (e.g., raise hand, peace sign).


Best Practices:

  1. Choose a Clean, Uncluttered Background  

    • Avoid busy or complex backgrounds. Plain, light-colored walls work best.

    • Remove visual distractions and background objects that could confuse the camera.

    • Avoid other people or motion in the background during pose-based tasks.
  2. Maintain Proper Camera Distance and Angle

    • Ensure the full upper body is visible (or full body if the activity requires it).

    • Keep the camera at chest or eye level with the child.

    • Avoid very close-up views that can distort body parts on screen.

  3. Ensure Good Lighting

    • Use natural light or soft indoor lighting.

    • Avoid shadows or overly dark environments that can hide body parts.

  4. Encourage Slow, Clear Movements

    • Fast movements may not be captured accurately, as the system processes frame by frame.

    • For activities involving multiple steps or poses, ask the child to pause briefly after each pose.

  5. Keep the Body Upright wherever Possible

    • Extreme tilts or bends of the head, torso, or limbs may result in missed detections.

    • If the activity involves bending or turning, guide the child to move slowly and return to an upright position afterward.

  6. Avoid Overlapping Limbs

    • Ensure that arms and hands do not block the face or overlap other body parts during actions.

    • Encourage poses where all limbs are clearly visible.


Gaze Detection: How to support clear gaze detection

Purpose: Tracks whether the child is looking on the screen to measure attention.


Best Practices:

  1. Make sure the child’s face is well-lit and clearly visible.
  2. Use bright, even lighting — avoid black or dark backgrounds.
  3. Ensure the child is centered in the camera frame. 
  4. Avoid reflections from glasses, if worn.
  5. Ensure the child is seated comfortably and remains still during gaze-based tasks.

Speech Detection: How to Support Clear Voice Input

Purpose: Detects verbal responses when prompted.


Best Practices:

  1. Speak Only When Prompted
    • The child should only speak when the speech icon appears, not during instruction playback. See speech icon: 

    • Do not speak on behalf of the child
  2. Use a Quiet Environment
    • Background noise (like TV, fans, or conversations) can interfere with speech detection.

    • Choose a quiet room with minimal echo for best results.

  3. Encourage Clear Speech
    • Ask the child to speak slowly and clearly. Encourage clear, single-word or phrase responses as required by the activity.

    • Avoid whispering or shouting — a normal speaking volume is ideal.

  4. Check Microphone Access
    • Ensure microphone permissions are enabled on the device and that the mic is functioning correctly.


Interactive Animation Games

Purpose: Engages the child through interactive stories and animation-based learning.


Best Practices:

  1. The child should wait for the instruction to finish before interacting.
  2. Avoid tapping or clicking as soon as the game loads.
  3. Let the audio play completely for better understanding.
  4. Encourage attention and patience — some animations are sequential and reactive.

 General Tips for All Detection Features

  1. Restart the app if detection features are unresponsive.

  2. Keep devices charged and ensure a strong internet connection.

  3. Stay nearby in early sessions to guide the child as they learn the flow.



Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article