Do Good Project Week 13 : Sound Editing & Technical Prototype ( Sensitivity & Timer )

Sound Editing

I start to work on the narration. To make the voice more echoing, I have added delay and reverb on the narration. 

Scripting

Critics

For the narration, my lecturer thinks that it's too forceful as I'm forcing the user to connect with Melda immediately when they step into the room. I should improvise the narration and avoid giving instructions to the users.

Besides, I also need to be creative for the audio aid to guide users to express freely in order to collect the input narration. I have to say things differently with the same intention in this context. I can feel that my lecturer is pushing me more to make improvements on this project. I can feel she is hoping me to create something that would really touch the user's heart and help them to think their mindset differently. For the next plan, I have to squeeze my brain and think critically for better improvements. 

Fabrication

I've added 4 strips of LEDs strips with 30 LEDs each to attach to the head. Besides, there are some wires that are broken during the experimentations. I went to fix it to make sure everything goes smoothly.

Problems

There are a few problems that occurred in the technical part when I'm sending serial data to the Arduino. I've found out something interesting about the color of the LED strips and this took me quite some time to figure it out. The codes are the same on two different sketches but the light effect couldn't work well differently. For example, light effect function no1 works well on sketch A, but light effect function no 2 doesn't work well. However, function no2 doesn't work well on sketch A, but function no2 works well on sketch B.

In my case, the CRGB white color of neopixels keeps showing yellow color instead of white. Although both sketches have exactly the same code, one shows white but the other shows yellow. Here is an example of it.

At last, I finally know the answer to it. This might be due to the low power supply as there are too many LEDs lights. It might not be able to support too many LEDs light and thus in return showing yellow colors. After I adjusted the brightness for this specific function, it works fine. 

Technical

Sensitivity

To increase the sensitivity of intonation, I've implemented the confidence level of the detection. With the written code, the effects will only change when it detects the intonation that is up to 95%.

Timer

I have added the timer to play the audio clip based on my previous exploration in Week 9. I've tried on millis, callback, setInterval, and setTimeOut. At last, I've decided to work on this method to set the timer. In this case, I've successfully coded out for Melda to change into white colors when she's talking, Afterwards, the function will change back into detecting the intonations after the narration clip was played finish. 

For my next plan, I need to add the ambiance of the ocean to the environment. However, I've forgotten to record the ocean ambiance as the background noise. This is the part where I need to start it out since the technical parts are mostly settled. 

Sources

jen_GSA(n.d) Teachable Machine: Image Classification + Sound Output. p5.js Web Editor. Available at: https://editor.p5js.org/jen_GSA/sketches/fYYmPX4g