Archive for the ‘electret microphone’ Category

Machine learning for the maker community

Wednesday, April 27th, 2016

mellis-aday

At Arduino Day, I talked about a project I and my collaborators have been working on to bring machine learning to the maker community. Machine learning is a technique for teaching software to recognize patterns using data, e.g. for recognizing spam emails or recommending related products. Our ESP (Example-based Sensor Predictions) software recognizes patterns in real-time sensor data, like gestures made with an accelerometer or sounds recorded by a microphone. The machine learning algorithms that power this pattern recognition are specified in Arduino-like code, while the recording and tuning of example sensor data is done in an interactive graphical interface. We’re working on building up a library of code examples for different applications so that Arduino users can easily apply machine learning to a broad range of problems.

The project is a part of my research at the University of California, Berkeley and is being done in collaboration with Ben Zhang, Audrey Leung, and my advisor Björn Hartmann. We’re building on the Gesture Recognition Toolkit (GRT) and openFrameworks. The software is still rough (and Mac only for now) but we’d welcome your feedback. Installations instructions are on our GitHub project page. Please report issues on GitHub. (more…)

Arduino Circuit Bending Workshop in Torino

Wednesday, November 2nd, 2011

Il prossimo week-end a Torino si terrà un workshop gratuito di Arduino (un kit opzionale potrà essere comprato per partecipare qualora non disponeste dei materiali elencati) sul circuit bending e la generazione di suoni con la scheda.

Un workshop di tre giorni per smontare riciclare e far suonare vecchi strumenti elettronici, creare una digital toys orchesta e sfilare in parata a Paratissima.

Il circuit Bending è una pratica molto diffusa tra gli sperimentatori musicali. Soprattutto sulla scena della musica elettronica sono sempre più frequenti gli artisti che si creano controller o addirittura strumenti musicali personalizzati.

Nel workshop saranno coinvolte diverse discipline: toy hacking, riciclo elettronico, elettronica di base, sintesi sonora, programmazione ad oggetti e faremo largo uso di Arduino per comandare i nuovi strumenti.

Il workshop è gratuito, a carico dei partecipanti il costo dei materiali e l’acquisto del kit-workshop.
maggiori informazioni quì!

Per partecipare registrati qui.

via [FablabItalia]

Samsa II, The Hexapod

Tuesday, March 29th, 2011

[pabloxid] shared an Hexapod project on the forum based on an Arduino MEGA 1280 and 18 Dynamixel AX-12 motors:

SAMSA is based on the Wiring board, with an ATmega128 microcontroller, and SAMSA II on the Arduino Mega, with an ATmega1280. Both are pretty similar, tough the ATmega1280 has 8 KB SRAM, twice the ATmega128. For SAMSA II the Arduino IDE was not used. The software was written directly in C++, using some libraries from both Arduino and Wiring.

SAMSA II has also two additional microcontrollers. One is an old Arduino Mini (ATmega168) located in the head, tasked with handling the sensors. The other is an ATmega8 and is integrated in the display. The firmware in the display was replaced with another one, freeing the main microcontroller from handling the display pixel by pixel, storing the frame buffer, etc.

The head’s microcontroller is responsible for sampling, filtering and processing sensor’s data. The data from the Sharp distance sensor and the lateral IR sensors are combined in a single “super smart distance sensor”. This microcontroller also decodes the data coming from the 38 KHz IR receiver, used for the Remote Control.

These two additional microcontrollers further reduce the load on the main microcontroller, allowing for more sophisticated behaviours.

(more…)

Arduino Realtime Audio Spectrum Analyzer with Video out! Arduino Realtime Audio Spectrum Analyzer with Video out! Arduino Realtime Audio Spectrum Analyzer with Video out!

Tuesday, November 16th, 2010

[Paul Bishop] shared code & pics about his project mixing a 8 bit FFT library found on the forum (in C) and the TvOut library.
The first piece- data collection- is fairly standard.  I use an electret microphone (which alone only produces a few mV output, far too low for our Arduino to use directly) with a transistor amplifier as the signal source, which is then sampled via the ADC on the Analog 0 pin of the Arduino.
To do spectrum analysis however, you need to capture signal over time, then process that data with what is known as a Fourier Transformation.  This magical process takes a signal and breaks it down into buckets  based upon frequencies found within the sample.  This produces a remarkably good picture of the signal.. and if displayed, functions as a visual spectrum analyzer if looped over and over.
This post contains a library which performs both the sampling and the Fast Fourier Transformation completely in C in 8 bits, amazing fast considering that fact, and uses a few tricks to be really stingy on memory, which is at a premium on Arduino- especially with the TVout data space eating up quite a bit.  Since the Atmega 328 only has 2k of RAM, every byte counts.  Matrix math done like this is nothing short of awesome.  Best of all, it’s usable as a library.  Cut and paste the .cpp and .h into a new folder named “FFT” in the Libraries directory.  My Arduino project code is adapted from the original code from the forum-posted Arduino program.
via [Blurtime]

 

[Paul Bishop] shared code & pics about his project mixing a 8 bit FFT library found on the forum (in C) and the TvOut library.
The first piece- data collection- is fairly standard. I use an electret microphone (which alone only produces a few mV output, far too low for our Arduino to use directly) with a transistor amplifier as the signal source, which is then sampled via the ADC on the Analog 0 pin of the Arduino.
To do spectrum analysis however, you need to capture signal over time, then process that data with what is known as a Fourier Transformation. This magical process takes a signal and breaks it down into buckets based upon frequencies found within the sample. This produces a remarkably good picture of the signal.. and if displayed, functions as a visual spectrum analyzer if looped over and over.
This post contains a library which performs both the sampling and the Fast Fourier Transformation completely in C in 8 bits, amazing fast considering that fact, and uses a few tricks to be really stingy on memory, which is at a premium on Arduino- especially with the TVout data space eating up quite a bit. Since the Atmega 328 only has 2k of RAM, every byte counts. Matrix math done like this is nothing short of awesome. Best of all, it’s usable as a library. Cut and paste the .cpp and .h into a new folder named “FFT” in the Libraries directory. My Arduino project code is adapted from the original code from the forum-posted Arduino program.
via [Blurtime]

[Paul Bishop] shared code & pics about his project mixing a 8 bit FFT library found on the forum (in C) and the TvOut library.
The first piece- data collection- is fairly standard. I use an electret microphone (which alone only produces a few mV output, far too low for our Arduino to use directly) with a transistor amplifier as the signal source, which is then sampled via the ADC on the Analog 0 pin of the Arduino.
To do spectrum analysis however, you need to capture signal over time, then process that data with what is known as a Fourier Transformation. This magical process takes a signal and breaks it down into buckets based upon frequencies found within the sample. This produces a remarkably good picture of the signal.. and if displayed, functions as a visual spectrum analyzer if looped over and over.
This post contains a library which performs both the sampling and the Fast Fourier Transformation completely in C in 8 bits, amazing fast considering that fact, and uses a few tricks to be really stingy on memory, which is at a premium on Arduino- especially with the TVout data space eating up quite a bit. Since the Atmega 328 only has 2k of RAM, every byte counts. Matrix math done like this is nothing short of awesome. Best of all, it’s usable as a library. Cut and paste the .cpp and .h into a new folder named “FFT” in the Libraries directory. My Arduino project code is adapted from the original code from the forum-posted Arduino program.
via [Blurtime]

 

Arduino Realtime Audio Spectrum Analyzer with Video out! Arduino Realtime Audio Spectrum Analyzer with Video out! Arduino Realtime Audio Spectrum Analyzer with Video out!

Tuesday, November 16th, 2010

[Paul Bishop] shared code & pics about his project mixing a 8 bit FFT library found on the forum (in C) and the TvOut library.
The first piece- data collection- is fairly standard.  I use an electret microphone (which alone only produces a few mV output, far too low for our Arduino to use directly) with a transistor amplifier as the signal source, which is then sampled via the ADC on the Analog 0 pin of the Arduino.
To do spectrum analysis however, you need to capture signal over time, then process that data with what is known as a Fourier Transformation.  This magical process takes a signal and breaks it down into buckets  based upon frequencies found within the sample.  This produces a remarkably good picture of the signal.. and if displayed, functions as a visual spectrum analyzer if looped over and over.
This post contains a library which performs both the sampling and the Fast Fourier Transformation completely in C in 8 bits, amazing fast considering that fact, and uses a few tricks to be really stingy on memory, which is at a premium on Arduino- especially with the TVout data space eating up quite a bit.  Since the Atmega 328 only has 2k of RAM, every byte counts.  Matrix math done like this is nothing short of awesome.  Best of all, it’s usable as a library.  Cut and paste the .cpp and .h into a new folder named “FFT” in the Libraries directory.  My Arduino project code is adapted from the original code from the forum-posted Arduino program.
via [Blurtime]

 

[Paul Bishop] shared code & pics about his project mixing a 8 bit FFT library found on the forum (in C) and the TvOut library.
The first piece- data collection- is fairly standard. I use an electret microphone (which alone only produces a few mV output, far too low for our Arduino to use directly) with a transistor amplifier as the signal source, which is then sampled via the ADC on the Analog 0 pin of the Arduino.
To do spectrum analysis however, you need to capture signal over time, then process that data with what is known as a Fourier Transformation. This magical process takes a signal and breaks it down into buckets based upon frequencies found within the sample. This produces a remarkably good picture of the signal.. and if displayed, functions as a visual spectrum analyzer if looped over and over.
This post contains a library which performs both the sampling and the Fast Fourier Transformation completely in C in 8 bits, amazing fast considering that fact, and uses a few tricks to be really stingy on memory, which is at a premium on Arduino- especially with the TVout data space eating up quite a bit. Since the Atmega 328 only has 2k of RAM, every byte counts. Matrix math done like this is nothing short of awesome. Best of all, it’s usable as a library. Cut and paste the .
cpp and .h into a new folder named “FFT” in the Libraries directory. My Arduino project code is adapted from the original code from the forum-posted Arduino program.
via [Blurtime]

[Paul Bishop] shared code & pics about his project mixing a 8 bit FFT library found on the forum (in C) and the TvOut library.
The first piece- data collection- is fairly standard. I use an electret microphone (which alone only produces a few mV output, far too low for our Arduino to use directly) with a transistor amplifier as the signal source, which is then sampled via the ADC on the Analog 0 pin of the Arduino.
To do spectrum analysis however, you need to capture signal over time, then process that data with what is known as a Fourier Transformation. This magical process takes a signal and breaks it down into buckets based upon frequencies found within the sample. This produces a remarkably good picture of the signal.. and if displayed, functions as a visual spectrum analyzer if looped over and over.
This post contains a library which performs both the sampling and the Fast Fourier Transformation completely in C in 8 bits, amazing fast considering that fact, and uses a few tricks to be really stingy on memory, which is at a premium on Arduino- especially with the TVout data space eating up quite a bit. Since the Atmega 328 only has 2k of RAM, every byte counts. Matrix math done like this is nothing short of awesome. Best of all, it’s usable as a library. Cut and paste the .cpp and .h into a new folder named “FFT” in the Libraries directory. My Arduino project code is adapted from the original code from the forum-posted Arduino program.
via [Blurtime]