The “Zap” Project
A “secret keeper” for kids age 5 and up
A “simpler” interactive talking toy that doesn’t just spit back whatever it hears, tries to teach a 5 year old how to count, strongly hints at “robots are always war machines” or records whatever it hears for storage at some corporation’s head office. The stuffed toy will be a friend and confidant to his companion child (of whatever age). It will listen to and “understand” human speech but is incapable of making enough of the same noises human speech consists of and has developed its own “language” called “Zap”. The child (or older person) will have to put the effort into learning to understand “Zap”. What this gives a child is a toy/companion/friend he or she can tell anything that will “keep his or her secrets” because nobody else in the house understands what the toy is “saying”.
This is intended to be a child’s “best friend” who listens to what he’s told and doesn’t tell “Mom and Dad” about everything. I believe this will stimulate “true” imagination (that which is triggered by nothing but itself) in the user by requiring the willing effort to imagine that “Zap” is actually speech and that there is an actual response from the toy. Preliminary sketch of the concept is below. The sketch is of a teddy-bear but the toy could be any stuffed animal of the necessary size.
General contemplation of what I want the project to do and what the difference is between this and other interactive toys.
Most of the “interactive” toys I’ve seen on the market have several drawbacks in my opinion:
- Pre-recorded and predictable human phrases triggered by a squeeze or similar
- Pre-recorded “babble” triggered as above
- Random and pointless movement (waddle or roll or other)
- Overly “educational” (teach my toddler to count, etc.) for my target age group
- Some of them actually record ambient (i.e. your conversations etc.) and store them at the manufacturer’s head office (privacy issues abound here see article copied from the New York Times below):“SAN FRANCISCO — My Friend Cayla, a doll with nearly waist-length golden hair that talks and responds to children’s questions, was designed to bring delight to households. But there’s something else that Cayla might bring into homes as well: hackers and identity thieves.Earlier this year, Germany’s Federal Network Agency, the country’s regulatory office, labeled Cayla “an illegal espionage apparatus” and recommended that parents destroy it. Retailers there were told they could sell the doll only if they disconnected its ability to connect to the internet, the feature that also allows in hackers. And the Norwegian Consumer Council called Cayla a “failed toy.”
The doll is not alone. As the holiday shopping season enters its frantic last days, many manufacturers are promoting “connected” toys to keep children engaged. There’s also a smart watch for kids, a droid from the recent “Star Wars” movies and a furry little Furby. These gadgets can all connect with the internet to interact — a Cayla doll can whisper to children in several languages that she’s great at keeping secrets, while a plush Furby Connect doll can smile back and laugh when tickled.”
- They’re some variation of the “remote control robot” (usually some type of “armed combat units”)
- Tremendously expensive for the actual product offerings
Online research for the parts that may be required outside of the class kit (preliminary list below):
- Feather Wing with MP3 (or other format) storage (internally or on a micro-SD card) and playback capability
- Found Adafruit Music Maker Feather Wing.
- There is, however no (detectable from the product description) recording capability. I may have to pre-record some speech, pre-“scramble” it into “Zap” and work with a variable timer (1 to 3 seconds at random).
- Found the Electret Mic Amp – MAX9814.
- I’m going to use the microphone as a trigger for a response rather than a pickup for ambient voices. There will be no privacy concerns as it will only trigger the playback.
- Found the LiPo 2000mAh #2011 battery
- Should be enough to power the toy as there will be no continuous drain.
Feather Wing, microphone and lithium ion battery on order from Adafruit (DHL delivery … hopefully fast).
Preliminary Fritzing Sketch:
Preliminary Schematic Diagram:
Found a toy at Value Village (see photo below).
This will become “Murgatroyd the Talking Moose”. It’s suitable due to the zippered opening in the back (making parts insertion/repair simpler for the prototype). Created a user survey form for testing period (link below).
Parts have arrived (yay). Unfortunately, I can’t get downtown to solder them today … will go tomorrow. Spent over an hour making and recording silly noises in separate tracks to load onto the SD card. This was simply taking a consonant and adding a vowel (Ba pronounced as “Bah” for example) or just a vowel alone. I intend to simply use the “shuffle” feature to make Murgatroyd “speak” in “Zap”.
All soldered up and ready to go (hopefully). Note from photo of the circuit assembly below that I’ve deliberately installed the header pins such that the “working surfaces” of the Feather and Feather Wing are facing rather than exposed. This will help prevent (hopefully) any jerked or pulled contacts and impact injuries inside the toy. Music Maker library downloaded and installed in Arduino. Files transferred to SD card as TRACK001 – TRACK037.
Adafruit’s basic test sketches for the Music Maker and microphone check (link below):
The code for the Music Maker verifies and uploads but I’m getting message “SD failed, or not present”. Tried re-seating the SD card with same result. I hope I don’t have what the Adafruit site defines as a “non brand knock-off”. Visually checked all solder points and connection points on the SD card … everything seems OK but will not recognize the card. Have tried two brands (Kingston and Nextech) 32 GB micro-SD cards. Adafruit’s test sketch for the microphone seems to recognize the input (getting a result in the serial monitor anyway).
I hate this … I absolutely hate it. Arduino’s test code for the microphone gives me a result I can use as input for the trigger and the Music Maker feather wing test code verifies and uploads but will not find the SD card (error copied below from serial monitor output):
Adafruit VS1053 Feather Test
SD failed, or not present
I’ve tried everything I can think of. Without access to the sound files I can’t code to test for the full project. Failure in coding, SD card, wiring or other. I’ve simulated the final assembly (all the pieces fit in the toy) but it doesn’t work (photo below)
Day Eight and Later (December 20 – 26):
Ran the user survey as best I could.
Eight survey responses (neighbours and relatives) about the toy (solely as concept since I can’t make the thing work).
Final Fritzing Sketch
Speculative Repair Kit List (If it actually worked):
- Micro-screwdriver set
- black electrician’s tape
- Needle & thread (for possible repairs to Murgatroyd)
- soldering kit (for possible repairs to the circuitry)
B.O.M. for the project: