Seeking advice on stereo audio transmission from a single headset device

[similingus]

I’m working with the Audio Application (from the Nordic sample program) on an nRF5340 Audio DK as a gateway, and a custom board with an nRF5340 as a headset. The custom headset board has two audio channels, and my goal is to transmit the stereo audio from the single headset to the gateway in the bidirectional mode.

I’ve made modifications to the Audio Application to enable the gateway device to decode two audio channels and the headset device to encode two channels. However, I'm currently facing an issue with the following error message on the headset device: “<err> bt_le_audio_tx: The encoded data size does not match the SDU size”.

My question is: Is it possible to achieve this stereo audio transmission setup by only configuring Kconfig and making minor changes to the code, or do I need to make more substantial modifications?

Any insights or suggestions would be greatly appreciated!

Thanks in advance!

NCS version: 2.6.1

[dejans]

Hi,

Can you show your modifications to the nrf5340_audio application?

Can you provide your full application log?

Best regards,
Dejan

[similingus]

Hi Dejan, thank you for your quick response. Here are the code changes and the log.

Code modifications:

// src/audio/audio_system.c

static void audio_gateway_configure(void)
{
    // ..
    sw_codec_cfg.decoder.num_ch = 2;
    sw_codec_cfg.decoder.channel_mode = SW_CODEC_STEREO;
    // ..
}

static void audio_headset_configure(void)
{
    // ..
	sw_codec_cfg.encoder.num_ch = 2;
	sw_codec_cfg.encoder.channel_mode = SW_CODEC_STEREO;
	// ..
}

 

// unicast_server/Kconfig.defaults
config LC3_ENC_CHAN_MAX
default 2

// unicast_client/Kconfig.defaults
config LC3_DEC_CHAN_MAX
default 2

//prj.conf
CONFIG_STREAM_BIDIRECTIONAL=y

 

// src/bluetooth/bt_stream/bt_le_audio_tx.c
int bt_le_audio_tx_send(struct bt_bap_stream **bap_streams, struct le_audio_encoded_audio enc_audio,
uint8_t streams_to_tx)
{
    // ...
	// if ((enc_audio.num_ch == 1) || (enc_audio.num_ch == streams_to_tx)) {
	// 	data_size_pr_stream = enc_audio.size / enc_audio.num_ch;
	// } else {
	// 	LOG_ERR("Num encoded channels must be 1 or equal to num streams");
	// 	return -EINVAL;
	// }
    data_size_pr_stream = enc_audio.size; // temporary adjustment
    // ...
}

Log: 

*** Booting nRF Connect SDK v3.5.99-ncs1-1 ***
-- [00:00:00.911,956] <dbg> main: main: nRF5340 APP core started
HL [00:00:01.065,979] <inf> fw_info:
nRF5340 Audio nRF5340 Audio DK cpuapp
NCS base version: 2.6.1
Cmake run : Tue Nov 19 10:50:14 2024
HL [00:00:01.065,979] <inf> fw_info: ------- DEBUG BUILD -------
HL [00:00:01.065,979] <inf> fw_info: HEADSET left device
HL [00:00:01.085,266] <inf> bt_mgmt: Clearing all bonds
HL [00:00:01.085,632] <inf> bt_mgmt_ctlr_cfg: Controller: SoftDevice: Version 5.4 (0x0d), Revision 8591
HL [00:00:01.085,845] <inf> bt_mgmt: Local identity addr: D9:D6:B4:D9:C1:D1 (random)
HL [00:00:01.087,463] <dbg> main: ext_adv_populate: Size of adv data: 48, num_elements: 6
HL [00:00:01.089,996] <inf> bt_mgmt_adv: Local addr: 65:0E:75:D0:2C:B0 (random)
HL [00:00:01.090,515] <inf> bt_mgmt_adv: Advertising successfully started
HL [00:00:01.211,242] <inf> bt_mgmt: Connected: 45:DD:13:98:7A:E0 (random)
HL [00:00:01.211,273] <inf> main: Connected
HL [00:00:01.933,288] <inf> main: Security changed
HL [00:00:04.093,139] <inf> bt_content_ctrl_media: Discovery of MCS finished
HL [00:00:06.092,864] <inf> unicast_server: LC3 codec config for sink:
HL [00:00:06.092,895] <inf> unicast_server: Frequency: 16000 Hz
HL [00:00:06.092,895] <inf> unicast_server: Duration: 10000 us
HL [00:00:06.092,895] <inf> unicast_server: Channel allocation: 0x1
HL [00:00:06.092,895] <inf> unicast_server: Octets per frame: 40 (32000 bps)
HL [00:00:06.092,895] <inf> unicast_server: Frames per SDU: 1
HL [00:00:06.093,353] <dbg> main: le_audio_msg_sub_thread: Received event = 1, current state = 1
HL [00:00:06.093,353] <dbg> main: le_audio_msg_sub_thread: LE audio config received
HL [00:00:06.093,353] <dbg> main: le_audio_msg_sub_thread: Sampling rate: 16000 Hz
HL [00:00:06.093,383] <dbg> main: le_audio_msg_sub_thread: Bitrate (compressed): 32000 bps
HL [00:00:06.093,383] <dbg> main: le_audio_msg_sub_thread: Direction: 1
HL [00:00:06.093,383] <dbg> audio_system: audio_system_config_set: audio_system_config_set Encoder: 0, Decoder: 16000, Bits: 0
HL [00:00:06.093,414] <dbg> audio_system: audio_system_config_set: audio_system_config_set Decoder sample rate: 0
HL [00:00:06.094,299] <inf> unicast_server: LC3 codec config for source:
HL [00:00:06.094,299] <inf> unicast_server: Frequency: 16000 Hz
HL [00:00:06.094,299] <inf> unicast_server: Duration: 10000 us
HL [00:00:06.094,329] <inf> unicast_server: Channel allocation: 0x1
HL [00:00:06.094,329] <inf> unicast_server: Octets per frame: 40 (32000 bps)
HL [00:00:06.094,329] <inf> unicast_server: Frames per SDU: 1
HL [00:00:06.094,665] <dbg> main: le_audio_msg_sub_thread: Received event = 1, current state = 1
HL [00:00:06.094,665] <dbg> main: le_audio_msg_sub_thread: LE audio config received
HL [00:00:06.094,665] <dbg> main: le_audio_msg_sub_thread: Sampling rate: 16000 Hz
HL [00:00:06.094,696] <dbg> main: le_audio_msg_sub_thread: Bitrate (compressed): 32000 bps
HL [00:00:06.094,696] <dbg> main: le_audio_msg_sub_thread: Direction: 2
HL [00:00:06.094,696] <dbg> audio_system: audio_system_config_set: audio_system_config_set Encoder: 16000, Decoder: 0, Bits: 32000
HL [00:00:06.094,696] <dbg> audio_system: audio_system_config_set: audio_system_config_set Encoder sample rate: 16000
HL [00:00:06.173,522] <dbg> main: le_audio_msg_sub_thread: Received event = 2, current state = 1
HL [00:00:06.173,522] <dbg> main: le_audio_msg_sub_thread: Set presentation delay
HL [00:00:06.173,553] <inf> main: Presentation delay 10000 us is set by initiator
HL [00:00:06.173,553] <dbg> main: le_audio_msg_sub_thread: Received event = 2, current state = 1
HL [00:00:06.173,583] <dbg> main: le_audio_msg_sub_thread: Set presentation delay
HL [00:00:06.173,583] <inf> main: Presentation delay 10000 us is set by initiator
HL [00:00:06.499,664] <inf> unicast_server: Stream 0x2000a304 started
HL [00:00:06.499,847] <dbg> main: le_audio_msg_sub_thread: Received event = 3, current state = 1
HL [00:00:06.499,847] <dbg> main: le_audio_msg_sub_thread: LE audio evt streaming
HL [00:00:06.499,877] <dbg> audio_system: audio_system_start: Init codec
HL [00:00:06.499,877] <dbg> audio_system: audio_system_start: audio_system_start Encoder sample rate: 16000
HL [00:00:06.499,877] <dbg> sw_codec_select: sw_codec_init: INIT LC3
HL [00:00:06.504,119] <dbg> sw_codec_select: sw_codec_init: LC3 encoder enabled
HL [00:00:06.504,150] <dbg> sw_codec_select: sw_codec_init: Encode: 16000Hz 16bits 10000us 32000bps 2 channel(s)
HL [00:00:06.504,455] <dbg> sw_codec_select: sw_codec_init: LC3 decoder enabled
HL [00:00:06.504,486] <dbg> sw_codec_select: sw_codec_init: Decode: 16000Hz 16bits 10000us 1 channel(s)
HL [00:00:06.510,375] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.520,385] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.530,364] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.532,379] <inf> unicast_server: Stream 0x2000a328 started
HL [00:00:06.540,374] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.550,384] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.560,363] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.570,343] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.580,383] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.590,362] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.600,372] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.610,351] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.620,361] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.630,371] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.640,350] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.650,360] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.660,369] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.670,349] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.680,358] <wrn> le_audio_rx: Not in streaming state, throwing data: 1
HL [00:00:06.680,603] <dbg> main: le_audio_msg_sub_thread: Received event = 3, current state = 0
HL [00:00:06.680,603] <dbg> main: le_audio_msg_sub_thread: LE audio evt streaming
HL [00:00:06.680,603] <dbg> audio_system: audio_system_encoder_start: Encoder started
HL [00:00:06.680,633] <dbg> main: le_audio_msg_sub_thread: Got streaming event in streaming state
HL [00:00:06.693,481] <err> bt_le_audio_tx: The encoded data size does not match the SDU size
HL [00:00:06.693,481] <wrn> main: Problem with sending LE audio data, ret: -22
HL [00:00:06.694,610] <wrn> audio_datapath: Data received, total under-runs: 12
HL [00:00:06.703,338] <err> bt_le_audio_tx: The encoded data size does not match the SDU size
HL [00:00:06.713,287] <err> bt_le_audio_tx: The encoded data size does not match the SDU size
HL [00:00:06.723,297] <err> bt_le_audio_tx: The encoded data size does not match the SDU size
HL [00:00:06.733,428] <err> bt_le_audio_tx: The encoded data size does not match the SDU size
HL [00:00:06.743,255] <err> bt_le_audio_tx: The encoded data size does not match the SDU size
HL [00:00:06.753,417] <err> bt_le_audio_tx: The encoded data size does not match the SDU size
HL [00:00:06.763,427] <err> bt_le_audio_tx: The encoded data size does not match the SDU size
HL [00:00:06.773,925] <err> bt_le_audio_tx: The encoded data size does not match the SDU size

[similingus]

Ok, thank you! I look forward to hearing the update.

[dejans]

Hi,

I have discussed the issue with our developers. Here is the demo branch
https://github.com/rick1082/sdk-nrf/commits/cis_stereo_audio_sink_pre280
Please let me know if you have any further questions.

Best regards,
Dejan

[similingus]

Hi Dejan, thank you for the quick update.

I’m not quite sure how to apply the modification to my code, as the base code differs from what you’ve provided.

I got the Audio Application project through the VS Code extension, nRF Connect. Specifically, I selected “Create a new application” > “Copy a sample” > “nRF5340 Audio application”. The toolchains and west workspace are v2.6.1.

Most of the lines of the code referenced in the commit do not exist in my codebase..

I also tried downloading the whole sdk-nrf/applications/nrf5340_audio directory in the cis_stereo_audio_sink_pre280 branch and compiling it, but ran into Kconfig warnings that prevented the build from succeeding.

Is the version I’m using no longer actively supported, or could you provide advice based on the code in that version?

[dejans]

Hi,

similingus said:

I’m not quite sure how to apply the modification to my code, as the base code differs from what you’ve provided.

Can you try to upgrade to the latest code base?

We will look into this issue, but it might take some time. I will get back to you hopefully by the end of this week.

Best regards,
Dejan

[similingus]

similingus said:

I also tried downloading the whole sdk-nrf/applications/nrf5340_audio directory in the cis_stereo_audio_sink_pre280 branch and compiling it, but ran into Kconfig warnings that prevented the build from succeeding.

I've tried the entire project in the branch you provided. Do I need to upgrade my SDK version? Also, how can I get the latest codebase?