@ejlane Yes, my speakers are best described as MQTT Text to Speech output devices. Basically, I wrote a small application which runs on a phone using MIT App Inventor. It subscribes to the Speak topic on my local MQTT broker. Whenever someone publishes to the Speak topic, it converts the text to speech. I have quite a few automations in Home Assistant which publish to Speak. I have 3 old phones set up as speakers around my house.

Regarding AI smart speakers, there is a lot of new stuff out there, with the release of tinyML.
From what I've read, the easiest type of speech recognition is "Keyword spotting" - see Edge Impulse
The next level is "Speech to Intent" - see Wio Terminal TinyML course
And the highest level is Large-vocabulary continuous speech recognition.

Hi everyone,
I'd like to share my latest project.
I've built up a node which monitors our washing machine for leaks and monitors our dryer for operation.
The node uses an arduino pro mini with an RFM69 radio. It runs continuously on a 5V wall wart supply regulated down to 3.3V.


The leak sensor is a classic probe type with 2 parallel conductors and a signal conditioner board (transistor based) which generates an analog voltage to represent the leak signal. When the signal exceeds the programmed threshold the leak is reported (using a binary sensor).


I used 14 gage solid copper wire to make the probe.

Dryer operation is sensed using an MPU-6050 accelerometer. I use the absolute value of the vibration signal and pass it through a moving average filter to get a steady signal.
The node is mounted to the dryer using magnets glued to the enclosure.

I also threw in a JDY-31 bluetooth module for serial monitoring/debug. So cheap and easy, why not?!

The vibration threshold can be set in Home Assistant, to allow for easy fine tuning.

I also added running statistics, updated every minute, of min (var1), max (var2) and average (var3) vibration levels to make it easier to set the threshold.

And to keep the wife happy , I have our home made smart speakers announce when the dryer is done.

@ejlane Oh, now I get your question. Text to Speech is built into the Android OS on the phone. There is a synthesis engine which runs on the phone and generates the speech. Although it likely has networked features, for example, loading different voices.
Project Alice looks quite interesting. I'll definitely be digging deeper. I see it runs on Raspberry Pi's, maybe it will run on my Ubuntu server too.

Thanks.
The motion sensor is used to turn on the lights in the garage via a Sonoff mini.
Also, I've had the garage door randomly open by itself (power glitches?) a few times over the years. So I'm adding an automation to close the door if no motion is present and it is after midnight.

@mfalkvidd Thanks. Yes those are spacers that snap on to the wires. I 3d printed them as well as the enclosure.

HI @Henil179

I made one myself a couple of years ago, and yes, it was a fair bit of trial and error to get it working.
Of course, the gain values will be dependent on a number of factors, for example, the torque constant of your motor, the gear ratio, the mass of the robot, where your centre of gravity is located, your loop rate, and where you've located your mpu-6050.
I located my mpu-6050 over the axis at 12 o'clock near the vertical centre of my bot.

What are you control variables?
I used pitch (angle of rotation around motor axis) calculated from gx, gz and tan function.
I also used pitch rate from the gyro.

Are you using the onboard DSP of the mpu-6050 to do the calculations?
I had no luck with this, the mpu-6050 kept freezing up, bad clone I guess.

You also have to verify the direction of rotation of your motor versus your control variables.
If you have the wrong direction, it will just run away.

My gains are Kp:41, Ki:160, Kd:0.4

I started with only Kp and kept increasing the value until the system became unstable, then I backed off to a stable point.
Next, I began increasing Ki. This will bring you to steady state stability. It should balance itself when Kp & Ki.
You will likely have to reduce Kp somewhat as you move to higher Ki values, if your system becomes unstable.
Finally, I added Kd, to help it react more quickly to changes in angle.

How do you change your gains?
I have a bluetooth module on mine, to allow me to send serial commands to the robot, to set the gains and other things.

My robot's biggest problem is gear backlash. This generates acceleration noise as it vibrates backward/forward, balancing itself. This vibration can feedback on itself and cause instability (when you increase Kp too high). It looks like the robot has Parkinsons when this happens!

I'd like to update my design to use a belt/gear drive or continuous rotation servo, but I haven't had a chance to get back to it.

I hope this helps.

Finished my Garage Door Node.
It has open/close magnetic reed switch sensors, a DS18B20 temperature sensor, a PIR motion sensor and a bjt to trigger the door to open/close/stop.

20210612_201851[1].jpg

@ejlane So far, I have a leak sensor under my clothes washer and hot water heater. I got the idea after my dish washer began leaking and damaged my kitchen cabinets. Dish washer is next...
I should clarify regarding the smart speaker. They really are MQTT text to speech speakers built from old cell phones I had laying around the house. I wrote an app in MIT app inventor. It could be expanded to play mp3 files of course and displaying images too.
I would like to create a smart speaker using Jetson Nano, but I haven't gotten around to it yet. Smart speakers might be an idea for a new thread. I haven't looked into it very deeply and could use some ideas too.

Documenting what I've found in the library code for the community.
When using #define MY_IS_RFM69HW:
The driver enables PA1 and PA2 and disables PA0 power amps. It automatically selects the right combination of PA1 and PA2 to achieve the target power level.
Power range is -2 to +20 dBm

When NOT using #define MY_IS_RFM69HW:
Then only PA0 is enabled.
PA0 is identical between the RFM69W and RFM69HW.
Power range is -18 to +13 dBm
Basically, your RFM69HW operates as an RFM69W.

Also, when using the new RFM69 driver, ATC mode is on by default, unless you explicitly turn it off using macro #define MY_RFM69_ATC_MODE_DISABLED
The default ATC target RSSI level is -80 dBm which you can change using #define MY_RFM69_ATC_TARGET_RSSI_DBM

Cheers

@kiesel Thanks for the schematic.
For the decoupling cap, it should be as close to the 3.3V & Gnd pins of the radio as possible.
Keep us posted, thanks.

@kiesel Looks like you are well on your way.
If you post a schematic, you'll likely get more feedback.
If you are using one of those Chinese step up converters to 3.3V, I found I had to add a 1 uF cap on Vin to get good performance at no load.

Most designs I've seen also have a 0.1 uF decoupling cap at the power connections to the radio.

I can't tell where you are connecting the door switch, directly to an unused pins on the pro mini?

One limitation to the pro mini is that you only have 2 interrupts and 1 is used for the radio, leaving you 1 for your project, which is probably fine for your application.
If you used a pro micro (32U4) instead, you would have 5 interrupts, leaving 4 for your project. And I don't think I've ever seen an rfm69 board for the pro micro, so it would be a first. Just a thought.

Hi @Leonel-Epps.
My pleasure, I've gotten a lot of good ideas from the Forum, its always good to give back.

P.S.
My washing machine leak sensor alerted me to a leak about 3 weeks. I checked, and sure enough, there was a small puddle of water under the machine. The door seal was wearing and had a few tears in the bellows.

I replaced the door seal (thank you youtube videos) and all is good. The washing machine is on the upper floor in my home and could have done serious damage to the lower level if it hadn't been detected quickly.

Regards,
KevinT

HI @Henil179

I made one myself a couple of years ago, and yes, it was a fair bit of trial and error to get it working.
Of course, the gain values will be dependent on a number of factors, for example, the torque constant of your motor, the gear ratio, the mass of the robot, where your centre of gravity is located, your loop rate, and where you've located your mpu-6050.
I located my mpu-6050 over the axis at 12 o'clock near the vertical centre of my bot.

What are you control variables?
I used pitch (angle of rotation around motor axis) calculated from gx, gz and tan function.
I also used pitch rate from the gyro.

Are you using the onboard DSP of the mpu-6050 to do the calculations?
I had no luck with this, the mpu-6050 kept freezing up, bad clone I guess.

You also have to verify the direction of rotation of your motor versus your control variables.
If you have the wrong direction, it will just run away.

My gains are Kp:41, Ki:160, Kd:0.4

I started with only Kp and kept increasing the value until the system became unstable, then I backed off to a stable point.
Next, I began increasing Ki. This will bring you to steady state stability. It should balance itself when Kp & Ki.
You will likely have to reduce Kp somewhat as you move to higher Ki values, if your system becomes unstable.
Finally, I added Kd, to help it react more quickly to changes in angle.

How do you change your gains?
I have a bluetooth module on mine, to allow me to send serial commands to the robot, to set the gains and other things.

My robot's biggest problem is gear backlash. This generates acceleration noise as it vibrates backward/forward, balancing itself. This vibration can feedback on itself and cause instability (when you increase Kp too high). It looks like the robot has Parkinsons when this happens!

I'd like to update my design to use a belt/gear drive or continuous rotation servo, but I haven't had a chance to get back to it.

I hope this helps.

@ejlane Your Death star speaker sounds pretty impressive! LEDs, timer/stopwatch, impact sensor, and of course speaker & microphone, she'll be loaded. You'll have to share a few pictures. How big will it be? Which Pi fits inside it?

@CrankyCoder The code can be found on Github: Washer-Dryer-Monitor

@CrankyCoder

1. I used a JDY-31 module without the carrier board. The module operates on 3.3V and has a header with 0.1" spacing, perfect for my needs.
2. Yes, I can share the code. There's the main sketch plus 2 libraries, one for the statistics and a modified version of the debounce library which allows me to debounce a boolean variable. The leak sensor needs this or you can get a flood of leak messages.
   Where do you want me to put the code? I can put it up on Github if you like. Might be a bit much to paste here.

Regarding the rhasspy project, I hadn't heard about this, I will definitely check it out. Thanks!

@ejlane Oh, now I get your question. Text to Speech is built into the Android OS on the phone. There is a synthesis engine which runs on the phone and generates the speech. Although it likely has networked features, for example, loading different voices.
Project Alice looks quite interesting. I'll definitely be digging deeper. I see it runs on Raspberry Pi's, maybe it will run on my Ubuntu server too.

@ejlane Yes, my speakers are best described as MQTT Text to Speech output devices. Basically, I wrote a small application which runs on a phone using MIT App Inventor. It subscribes to the Speak topic on my local MQTT broker. Whenever someone publishes to the Speak topic, it converts the text to speech. I have quite a few automations in Home Assistant which publish to Speak. I have 3 old phones set up as speakers around my house.

Regarding AI smart speakers, there is a lot of new stuff out there, with the release of tinyML.
From what I've read, the easiest type of speech recognition is "Keyword spotting" - see Edge Impulse
The next level is "Speech to Intent" - see Wio Terminal TinyML course
And the highest level is Large-vocabulary continuous speech recognition.

@ejlane So far, I have a leak sensor under my clothes washer and hot water heater. I got the idea after my dish washer began leaking and damaged my kitchen cabinets. Dish washer is next...
I should clarify regarding the smart speaker. They really are MQTT text to speech speakers built from old cell phones I had laying around the house. I wrote an app in MIT app inventor. It could be expanded to play mp3 files of course and displaying images too.
I would like to create a smart speaker using Jetson Nano, but I haven't gotten around to it yet. Smart speakers might be an idea for a new thread. I haven't looked into it very deeply and could use some ideas too.