I just searched about an open-source library to use of my project to build a voice controlled robot and found CMUsphinx. but my robot will build on ARM based microcontroller and I have not Linux or Windows on it. Is it possible to use CMUsphinx on ARM based microcontrollers? How can I do it?
Is there any better options instead of using CMUsphinx on ARM uControllers?
Actually, you can use a variant of CMUSphinx, known as PocketSphinx, on ARM devices. Most resources focus on Raspberry Pi, as it's a very popular ARM target.
The instructions for installing on a Raspberry Pi are here:
http://cmusphinx.sourceforge.net/2016/06/should-you-select-raspberry-pi-3-or-raspberry-pi-b-for-cmusphinx/
There's a voice assistant for Raspberry Pi that uses CMUSphinx:
http://cmusphinx.sourceforge.net/2014/04/jasper-personal-assistant-for-raspberry-pi/comment-page-1/
And, there's a recent blog post about running on the Raspberry Pi here:
http://cmusphinx.sourceforge.net/2016/06/should-you-select-raspberry-pi-3-or-raspberry-pi-b-for-cmusphinx/
Is it possible to use CMUsphinx on ARM based microcontrollers?
No, it's not possible to use CMUSphinx on microcontrollers.
but my robot will build on ARM based microcontroller
Consider using more powerful CPU, it will help you with other tasks like navigation or network sync as well.
Related
I am currently learning how to emulate both mips(el) and arm architecture using Qemu. The images which I used for both architecture is by aurel32 from this website, https://people.debian.org/%7Eaurel32/qemu/
I noticed that the mips images does not have initrd while the arm images have initrd. I am unsure why arm requires initrd when running Qemu but mips does not.
Anyone know why this is the case?
Thanks for the help!
There is no inherent reason -- you can build an Arm kernel with enough devices built-in to not need an initrd, and you can build a MIPS kernel that does require an initrd. You'd have to ask Aurelien why the two sets of images he built had different configs (assuming he even remembers 6 years later!), but I suspect it's just that for MIPS he was able to use the stock Debian kernel whereas for Arm he had to compile his own kernel, and so the configuration choices for the two weren't identical.
I've decided to make the step up from using PIC to ARM Cortex-M but having trouble finding an equivalent to the PICKIT to program it.
After googling the net I have been given loads of options but have no idea what to buy.
Can anyone out there give me any advise on what to buy?
I'm so used to buying a PIC16F microcontroller, using MPLAB and the PICKIT 3 I'm unsure about what I need.
I like the ST-Nucleo and the slightly more complex ST Discovery. They are inexpensive (start at around $13), versions for just about all STM32 series, built-in ST-LINK debug pod, and have Arduino style header connectors.
You can program it using free or commercial IDE/compilers, mBed compilers, and even the Arduino IDE. You can start with using the free CubeMX graphical generator to generate initialization code.
So literally all you have to spend minimally is just a $13 kit and a USB cable to get started.
I am new to Xilinx Zynq SoC. Zynq has ARM(dual cores). I am curious if it is possible to run program C/C++ program only on ARM processors without using the FPGA fabric. My research could not helped much.
IF yes, I would like to know some basic steps that I need to follow to achieve the goal.
Yes, you can run your own application code without using the FPGA fabric.
You can download the xilinx SDK http://www.xilinx.com/tools/sdk.htm and start by trying some of the hello world projects available. Adding a FPGA bitstream is optional.
Follow the Tutorial by Xilinx.
http://www.xilinx.com/support/documentation/sw_manuals/xilinx14_3/ug873-zynq-ctt.pdf.
It is clearly stated in section 5.2.4.
Thanks.
i coded a java applications that takes the pictures of a webcam that is attached to the pc. For that i use the lti-civil java-api. It works fine! This api uses a shared library -libcivil.so- for jni access to several native libs. The libcivil.so exists for linux 32/64bit osx and windows. But i want the application to run with raspberry pi which has arm-architecture.
My question is: is it possible to recompile the libcivil - shared lib for arm-architecture? iam not familiar with c++ so i don't know if it is possible in any way.
i found a way to obtain webcam images with java on a raspberry pi. I use the v4l4j library. It's a jni wrapper for the v4l2 api. It handles direct access to the /dev/video interface under linux. There is a way to compile it for the raspbian linux-port and it works great. But the raspberry seems to be too slow (5-8 fps and hight latency - about 300-500ms). But this is another thing. Now iam waiting for the hackberry or gooseberry... =)
compile v4l4j on raspberry pi how-to:
https://code.google.com/p/v4l4j/wiki/GettingStartedOnRPi
greez
daniel
The question below was related to the 8051 family as it has only the serial ports. Now when i found out that the ARM processors have inbuilt ethernet facility i decided to mov on to the ARM's. So which would be the Most easiest one to start of and as iam aware of keil environment i would like to work on anARM which supports keil .
Thnxx,
Cheers.
The TI (formerly Luminary Micro) parts are easy to use. The eval kits are cheap and there is a lot of documentation on the Luminary Micro web site, including sample code. I've been using the LM3S6965 EVK for an ethernet project. The eval kits can be ordered with Keil, IAR, Codesourcery or Code Red compilers.
Check out the AT91SAM7X.
AT91 SAM series from Atmel is really popular between hobbyists and amateurs, you will find a lot of open source examples and excessive topics regarding this CPU's. And yes, they are widely supported in Linux community.
Check this out: http://www.at91.com