Developing a MIC feature based on nRF54L15 LE Audio by I2S.

[LiYang1350305272]

nRF54L15  LE Audio with I2S interface, LC3 codec point-to-point audio
The specific flow is as follows:
The nRF54L15 communicates with the mic via I2S to capture audio data, performs LC3 encoding by LE Audio protocol.
May I ask if the example ncs\v3.0.2\zephyr\samples\bluetooth\bap_unicast_server, which transmits audio data to Window PC, can meet the above function?
This example is based on nRF52840 — is it possible to port it to the nRF54L15? And whether such MIC device could be recognized by Windows sytem？

[Elfving]

Hi,

That should be possible, as the nRF54L LE Audio. Though note that it does not have an AudioPLL, though for this use-case that is probably ok. 

May I ask if the example ncs\v3.0.2\zephyr\samples\bluetooth\bap_unicast_server

Yeah I believe that should work, though I would rather recommend sticking to the Audio samples from nordic. Those are better supported by us. What is available in regards to audio functionality might be a bit "bleeding" edge, so that what is supported in a zephyr sample might not match what the nordic SDC supports, meaning you also might have to use the zephyr controller. This sample might also need some extra configuration to work for a given nRF, though that is the case with all samples directly from zephyr.

And whether such MIC device could be recognized by Windows sytem？

I am not a 100% sure about what is needed for Windows to recognize it is a mic device, though similar things have been possible with the nordic sample for LE audio alongside the walkie talkie demo and the audio DK. So I assume this is possible. If you want me to double check I can ask the relevant R&D team.

Regards,

Elfving

[LiYang1350305272]

Hi Elfving,

Thank you for your reply! Please double check that porting ncs\v3.0.2\nrf\applications\nrf5340_audio\unicast_server to ncs\v3.0.2\nrf\applications\nrf_desktop to achieve MIC function is more available right? Do you have relevant porting documents or suggestions for reference to help me  shorten development duration?

Please double check with your R&D team that whether ncs\v3.0.2\nrf\applications\nrf5340_audio\unicast_server  could be recognized as LE Audio Device by Windows.

(Process : Digital MIC communicate with nRF54L15 DK by I2S, nRF54L15 send audio by LE Audio to Windows )

Look forward to your reply!

[Elfving]

LiYang1350305272 said:

(Process : Digital MIC communicate with nRF54L15 DK by I2S, nRF54L15 send audio by LE Audio to Windows )

Just to be clear, I assume you mean the device should send the LE Audio to a separate other device, that then streams the data over eg. UART or something to the windows computer? Or are you expecting Windows to support LE Audio?

I can tell you right now that the if the OS and HW in question supports LE Audio (eg a Windows computer) then it will be able to get the data streamed by an nRF54L15. Though I am not sure if Windows currently supports this. What I thought you were asking about is what is required for a connected nRF to be interpreted as something like an external mic or headset type device.

Regards,

Elfving

[LiYang1350305272]

Hi Elfving,

Not like this. nRF54L15 with MIC interfacing I2S send audio to Windows. 

Do you have  documents or recommended examples that porting ncs\v3.0.2\nrf\applications\nrf5340_audio\unicast_server to ncs\v3.0.2\nrf\applications\nrf_desktop?

[Elfving]

LiYang1350305272 said:

Not like this. nRF54L15 with MIC interfacing I2S send audio to Windows. 

I see. I thought Windows didn't support LE Audio yet, but it looks like it does! Then it should be fine.

LiYang1350305272 said:

Do you have  documents or recommended examples that porting ncs\v3.0.2\nrf\applications\nrf5340_audio\unicast_server to ncs\v3.0.2\nrf\applications\nrf_desktop?

Unfortunately we do not have any recommended ways to mix certain samples, I guess the best way would be the get a certain understanding of the samples and port one part of them at the time. Or essentially understand one, in order to port it to the other.

Though when it comes to the audio samples and nrf_desktop though, it might be a bit harder as they are both a bit complex. Nrf_desktop could be set up easier, but is currently created to also demonstrate CAF - which can look complex if you're not used to it. 

Getting a certain understanding of the samples would in any case be the first step.

Regards,

Elfving