Wondering if this codec is upgradeable.
Is it possible to include any modifications at the software level?
If not can you advise any device that's upgradeable?
Thanks, Pedro Silva.
What do you mean by "upgradable"? Any of the software that is in the device, like ASRCs and decimation filters are all hard coded and cannot be changed.
If you want something that can be changed then look at a DSP. But even the SigmaDSPs have parts of the device that are hard-coded.
So what would you want to upgrade on a codec, so I have an idea of what to suggest?
Thanks for the prompt feedback.
This is a customer question and this has to do with the fact that "tomorrow" it migh happen that there´s a new algorithm, much more efficient, doing noise/echo cancelling and voice recognition.
If this device would allow an upgrade this could be interesting.
The DSP can be a solution if it can be almost "off the shelf".
Well do we have a solution for them! They must look at the SigmaDSP line of products. The ADAU1442 and the ADAU1452 are the most capable parts we have at the moment. They can be programmed using SigmaStudio which allows for graphical programming of the DSP. The program can control the hardware for easy development and then the program can be loaded into an EEPROM so the DSP would boot itself up. Then any future adjustments to the program is simply done by loading in a new EEPROM image.
Noise and Echo cancelling has been done. Voice recognition is something I do not know much about. mmmike could help with some advice there.
If they want to have an off-the-shelf solution then perhaps they would want a third party development firm to deliver the application? It would also be good to know exactly what they want to accomplish? What are the details of the application?
Here is a link to the SigmaStudio page: SigmaStudio | product info The SigmaStudio | Analog Devices
When I say voice recognition I mean that voice signals should be acquired-digitalized-processed in order to be sent.
Maybe there are typical codecs that can handle the errors resulting from the digitalization process, codecs that somehow can recognize patterns...
Maybe Mike can give us a help here.
The actual solution is a DSP.
If you wanna go deep in this application we can talk privately.
Have a nice weekend.
Thank you. Pedro Silva.
Wish you a Happy 2016.
I would like to receive a feedback from mmmike in what concerns voice recognition but I was not able to find a way to reach him.
Can you help?
Retrieving data ...