Multiple BLE audio source(Mic) with single BLE audio sink

[chethan26]

Hello,

we are looking to develop multiple BLE audio source which sends the mic data to the receiver and the receiver receive the audio data from multiple transmitters over BLE, output through I2S.

Can anybody please help us how to implement with NRF53.

Similar Thread:

devzone.nordicsemi.com/.../multiple-ble-audio-source-devices-outputting-to-a-ble-audio-sink-device

Best Regards,

Chethan kumar

[Hung Bui]

Hi Chethan, 

Please give us more information about the application and the technology you plan to use: 

- Is it half duplex ? Meaning the receiver only receive and won't send any data back ? 

- What's the audio quality requirement  ? 

- How many stream are you planning to do ? 

- Are you thinking of Bluetooth LE Audio or you are looking at normal Bluetooth LE ? 

If you are doing normal BLE , you can think of what you are doing is simply sending compressed audio packets from multiple link and then uncompress and play them via I2S. 

[chethan26]

Hi Hung Bui,

Thank you for your response.

- Yes, it is half duplex. one receiver should receive audio data from multiple(4) transmitter. 

- This is for conference application and used like lapel mic.

- 4 transmitter 1 receiver.

Thinking of Bluetooth LE audio to get good audio quality but normal Bluetooth LE also fine. 

Can we use NRF53_Audio example to achieve this concept?

If yes, in CIS mode one gateway(Tx) and 2 headset(Rx). can we interchange to 2 Tx and 1 Rx?

Best Regards,

Chethan kumar

[ovrebekk]

I shared it by PM. Sorry for the delay....

[chethan26]

Thank you

[chethan26]

Hi Torbjørn,

As discussed with Gaurav, please update if you get any information on this requirement.

Best Regards,

Chethan kumar

[Gaurav Kapoor]

Hi Chethan

Since you want to transfer audio from four microphones to a central device. This is not something that we in the project have worked yet.  It is also outside the current official specifications for the controller, which is limited to two CISes.  So we can not guarantee that this will work, and we can not officially support it (yet). This may be something that we will officially support in the future, but we have no committed plans for that yet. That being said, the scenario is reasonable, as long as the bitrate and the number of retransmissions is kept within what radio airtime and scheduling allows (sorry, no concrete limits).  Extending the application to add two more devices is your scope of work for now. Unfortunately we cannot help more on this much and All the best for you to try out

tks

Gaurav

[chethan26]

Hi Gaurav,

Thank you for your response. 

Right now we are working with 2 microphone and succeeded by receiving mic data from 2 servers to client.

The receiving mic data packets are merging since we have single receiving buffer in this example, hence we are getting noisy audio out. If we know how use independent decoder for each mic, hope will succeed with 2 mic.

Best Regards,

Chethan kumar

[ovrebekk]

Hi 

I didn't realize it was possible to merge compressed LC3 data? But then again I don't know the codec specifics that well. 

Looking at the sd_codec_select.c implementation it seems it already supports decoding two parallel channels, when running the module in stereo mode. Please have a look here:
https://github.com/nrfconnect/sdk-nrf/blob/main/applications/nrf5340_audio/src/audio/sw_codec_select.c#L151-L183

Wouldn't you be able to do something similar?

Essentially you can run the sw_codec_lc3_dec_run(..) function as many times as you like (only limited by available CPU time), and provide separate buffers. 

Best regards
Torbjørn

[Selvaraj]

Hi,

I would like to record audio in mobile phone via BLE? Does Thingy 53 support BLE audio transfer to smart device from microphone in Thingy53 ?

If yes where I could find the example for the same both firmware and the app (android or iOS).

Best Regards,

Selvaraj 

[ovrebekk]

Hi Selvaraj

The Thingy:53 doesn't support this, no. For LE Audio testing it is recommended to get the nRF5340 Audio DK. 

If you have more questions about this please open a new ticket. 

Best regards
Torbjørn

[SNJY]

Is there a working example/demo for this scenario now? I am working on a similar use case and wanted to know if this is implemented.