Can anyone explain me how audio codecs work, how to select the best audio codec chip. How to interface it with a processor(Nvidia TX2).
Your question is very general thus making it a little difficult to answer. I can start by saying that the choice of a codec is a bit complicated because all of them are not suited for all applications. All of them may do the basic job but fall short due to other design constraints. Design constraints will be the usual:
Power requirements, either available power supply voltages and/or power consumption.
Sample rate requirements
Clocking details, available clocks, which device is master or slave, etc.
So it is kind of like choosing the correct bottle of wine for dinner. Lots of factors.
A codec is a combination of one or more ADCs (Analog to Digital Converters) and one or more DACs (Digital to Analog Converters). There are normally at least two of each and often the count of ADC vs DAC is not the same.
So there is an efficiency in having them all reside in one part so they are very common.
So let us know some of your application needs so we know what some of your design constraints are.
Also let us know what you plan on using it for?
I did take a quick look at the TX2 and it lists that it does have I2S ports. I do not know how many and which direction or what sample rates they can operate at.
I2S is a serial port designed for transmitting audio. I2S is actually a standard for stereo audio transmission with the clocks and data formatted a certain way. Some devices can use these serial ports in TDM mode. This is Time Domain Multiplexing. This is where more than two channels can be transmitted during a single sample period frame. If this TX2 can operate in TDM mode then you can get a large number of channels in and out. If it does not use TDM, then you need one I2S data port for each pair of channels.
All of our codecs will operate in I2S and TDM modes.
So give us some more detail and we can continue the conversation.
Retrieving data ...