Hello,
In our setup, we are working on Davinci DM6467 based board.Kernel running underneath is 2.6.32-rc2.For DSP side applications, we are using DVSDK_3_10_00_16.
In the application, analog signal is read from FPGA over HPI port. A LIVE analog TV signal broadcast is fed as an input to the board. The signal is further demodulated and decoded using ak4141 stereo demodulator.
Now, Sometimes, Bilingual programs are broadcast on the Television.
In the application ,ioctl is called which detects the bilingual/stereo/mono channels etc.
After determining that the channel being broadcast is a bilingual channel, i am looking forward to separating the left and right channel data and derive the 2 languages data separately(1 in left channel, and other in the right).
Firstly, is that possible..??
Later, i generate left and right channel audio samples in the .aac format, and expect that both would have different languages when bilingual program is on air.Am i going in the right direction??
Currently, with the following code, i am getting both aac files to be of the same language.
Following is what i have done to separate the left and right channel data from the data read over HPI port.
The code goes something like this:
/*L-Channel data*/
LchBuffHndArr = Buffer_create(ANALOG_BUFFER_SIZE, &bAttrs);
hpiBufPhyAddr.val1 = (unsigned int)Buffer_getPhysicalPtr(LchBuffHndArr);
/*R-channel data*/
RchBuffHndArr = Buffer_create(ANALOG_BUFFER_SIZE, &bAttrs);
hpiBufPhyAddr.val2 = (unsigned int)Buffer_getPhysicalPtr(RchBuffHndArr);
ioctl(fd_Fpga,
CAPTURE_I2S_DATA /*macro from davinci_hpi driver*/,
(void*)&hpiBufPhyAddr);
//this ioctl actually reads data from the FPGA over hpi port.
And, then read the data using user space pointers.
/*Read L-Channel data*/
LchDataBuf = (char*) Buffer_getUserPtr(LchBuffHndArr);
What can be done to retrieve 2 separate language's audio samples?
Thanks and regards,
Prathamesh Ghanekar
