No audio when streaming

[warrenaken]

Hi, Great Project!
I have managed to set up the XIAO-ESP32-C6 with the ICS43434 mic. I can access the new Web UI and confirm everything is in order, connecting to a client and streaming. When I use the RTSP stream in Bridnet-go I can see Total MB and Rate (KB/s) incoming but no audio. When testing in VLC I can see Codec: PCM S16 BE (s16b), Mono Channel, Sample Rate: 48000 HZS and 16 Bits per sample. Additionally under statistics I can see 'blocks' and buffering under audio and Demuxed data size, Content bitrate. However there is no sound.
My assumption is that I am 99% there but something minor is not allowing the sound to come through.
Any suggestions would be greatly appreciated.

[nate0402]

Were you able to find out what the issue was? I am having the same problem. The odd part is that It was working last night when I went to bed. Woke up this AM to show the family and now its not working. If I disconnect the 3v3 pin while streaming I get a blip when reconnecting (short static) in VLC, so I can tell that it is streaming. Seems like a wiring issue between ESP and ISC43434 OR a software setting. I have checked the wiring countless times.

[warrenaken]

Unfortunately not, i also thought it must be a hardware issue, replaced the wires and ordered a new mic. Same issue persists.

[truthringr]

I appear to be having the same issue. Did either of you get it resolved? I tried two different ESP32s, multiple mics (ICS 43434 and INMP 411) but no joy. It seems to be streaming fine but Birdnet-Go shows it as "silent."

[dalogue1]

[Sukecz]

This looks a bit different from your issue. In your case the stream seems to be alive — BirdNET-Go/VLC can see incoming data, bitrate and the correct audio format — but the actual audio samples appear to be silent or almost silent.

[Sukecz]

I do not want to say it is 100% a hardware issue, because I cannot reproduce it on my side, but this does point more toward the I2S input side: wiring, microphone channel selection, power, soldering, or a pinout difference on the microphone breakout board.

The firmware reads the left I2S channel, so the L/R or SEL pin should normally be connected to GND for the left channel. If it is connected to 3.3 V, or left floating, the microphone may output on the other channel and the stream can look valid but still be silent.

I tested the current version of the code on my setup and audio works correctly, so I am not currently able to reproduce the issue. That said, I am open to checking it further if you can share a clear wiring photo, exact ESP32 board, exact microphone module.

[warrenaken]

Thanks for the reply. I am using the speed studio Model:XIAO-ESP32-C6 (FCC ID:Z4T-XIAOESP32C6), with the ICS43434 I2S Mic. I have built 2 of these and swapped components around to test hardware. Also had someone check the solder for my sanity.
The pin out is shown below:

[Sukecz]

If the stream is connected but the audio is silent, please also check the microphone hole. It is possible that some solder flux or residue got into the acoustic port after soldering and blocked the microphone.

Please inspect it carefully and try to clean it gently if needed. Do not push anything sharp into the microphone hole.

[truthringr]

I'm not sure how helpful I can be to this conversation, but I just wanted to add that I had a very similar issue. I even tried multiple ESP32 units and microphones.

I did solve it, in the end, but I legitimately have no idea what caused the issue to resolve. I decided to try again on a completely different esp32 device (Elegoo esp32s wroom) with two ground pins to rule out poor wiring. I installed a really tiny app to just monitor the level input. I successfully got audio on one channel. Then moved on to wiring up another seeed studio xiao esp32-c6. Installed the audio monitoring program, and once again was able to get audio on one channel. Finally, tried flashing esp32-birndet-mic on there, and this time it worked.

I legitimately have no idea what changed. I had wired it up multiple times, with multiple different mics, tried switching left/right (via switching between 3V3/GND for the shared pin), tested with a voltmeter, etc. I'm assuming it was user error, unless there's something different with the way that I flashed it the final time. I'm at a loss!

[DervMan]

I'm having the same issue. I have two of these setups, one is still working perfectly but the other is silent. I've bought and installed a new mic board but the same problem. I've checked & replaced all the cables but silence persists. I'll have to order another ESP32-c6 and see what happens. I really curious to know where the problem is 🤔

[Sukecz]

I don't think the ESP32 itself is the problem. I would suspect the microphone board instead.

One thing that can happen during soldering is that flux gets into the tiny microphone opening. If that happens, the microphone can become almost completely silent, which would match exactly what you're seeing.

So before replacing the ESP32-C6, I'd definitely check the mic opening closely and, if possible, try another microphone board that hasn't been soldered yet.

[dalogue1]

Actually, I have had two different instances where a running/working setup where the sound stopped as you described. (streaming continues but becomes silent.) It absolutely can happen to a ICS43434 without any known reason.

I do not know if it is due to moisture due to condensation, a very small insect, or something else.

In both instances, I ONLY replaced the ICS43434 and it started working again.

[ejwolski]

I am having the same issue (no sound) using a XIAO ESP32-C3 board with the ISC43434 mic. The ESP32 serial monitor shows it is starting up successfully and that the client is connecting. BirdNET-Go shows the stream is connected and OK, but no sound. I will order another mic - do the ISC43434 mics have QC issues? Thanks!

[ejwolski]

I am still having the same issue on 1.23 as well as 1.22 - For round #2, I used a brand new XIAO ESP32-C6 plus a new ISC43434 mic. Per the logs on both BirdNet-GO and the ESP32, the unit is streaming but I am getting no audio whatsoever. I have two more ESP32-C6s on hand, but I don't think the ESP is the issue. I bought a couple of INMP441 mics to try as well.

I have double checked my wiring, and it appears correct per the diagram. I've also checked the mic, and I don't see any obstructions, flux, or other foreign material in the hole.

My plan is to do as mentioned above and breadboard this to try out the mics using I2S example code first. I love the idea of this wireless mic for BirdNET-Go, so I will definitely spend some more time troubleshooting.

UPDATE: Adafruit says the ICS43434 is discontinued, so I've ordered the SPH0645LM4H breakout. I saw that this is supported in the 1.23 release notes.

[Sukecz]

Thank you for the detailed update. I have not been able to reproduce this problem in my own testing, and there is currently no confirmed evidence that it is caused by the firmware.

Since the RTSP stream connects but remains silent even with a new ESP32-C6 and another ICS-43434, please try the new SPH0645 microphone when it arrives. With firmware 1.23, select Adafruit SPH0645 (MSB / left-justified) under Microphone format.

Please let us know whether the SPH0645 produces audio. That result should help distinguish a microphone-specific problem from something elsewhere in the setup.

[ejwolski]

I am pretty confident it's not firmware either. I've tripe checked my wiring - wires are short and to the correct pins on both the ESP32 and the mic. I should get my SPH0645 this weekend and will drop a note with results. Thanks for the response!

[ejwolski]

OK - success with the Adafruit SPH0645! I desoldered the ICS43434 and attached the SHP0645 since the pinouts were the same. As soon as I booted up the device, I was able to configure it in the web UI for the SHP0645 and successfully test it in BirdNET-Go (same as the ICS43434 since it's testing the RTSP connection and not the audio). Now for the real test...

Since it is now dark out and quiet, I played a bald eagle call on my iphone and got an immediate recognition in BirdNET-Go. Success!!!

I will order another SHP0645 mic for my second wireless microphone. Not sure what's up with the ICS43434s I have - maybe QC issues? I made no changes to anything other than swapping out the mic boards.

Cheers!

[warrenaken]

Thanks all for the troubleshooting. I have ordered the SPH0645 and will try that.

[ejwolski]

For what it's worth, here's my setup - a 3D printed case from Etsy (I think it was designed to be a Flock camera detector case). The seller was nice enough to make the buzzer cavity more of a shallow standoff for the mic board.

[dalogue1]

Can you please send the link to the ETSY case?

…

On Mon, Sep 21, 2026 at 3:58 PM Gene Wolski ***@***.***> wrote: For what it's worth, here's my setup - a 3D printed case from Etsy (I think it was designed to be a Flock camera detector case). The seller was nice enough to make the buzzer cavity more of a shallow standoff for the mic board. Image: IMG_5444 (view on web) <https://github.com/user-attachments/assets/6aec70b1-43e8-48a6-850f-3c7f27c5277e> — Reply to this email directly, view it on GitHub <#21?email_source=notifications&email_token=A7XGRLHHUTGOH4MRTE2ZXKL5QGI63A5CNFSNUABIM5UWIORPF5TWS5BNNB2WEL2ENFZWG5LTONUW63SDN5WW2ZLOOQXTCOBVGQ2TAOJQUZZGKYLTN5XKOY3PNVWWK3TUUVSXMZLOOSWGM33PORSXEX3DNRUWG2Y#discussioncomment-18545090>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7XGRLD3YAY4KWYVBMQWUI35QGI63AVCNFSNUABIKJSXA33TNF2G64TZHMYTANBZGI2TENJTGQ5UI2LTMN2XG43JN5XDWOJWHA2TAMRTUF3AE> . You are receiving this because you commented.Message ID: ***@***.*** com>

[ejwolski]

Certainly!

https://www.etsy.com/listing/4465175905/seeed-xiao-esp32s3-case-petg-cf?ls=r&external=1&ref_group_id=hp_content_grouping_gqc_v3__326983487841&ref=hp_content_grouping-1-1&sr_prefetch=1&pf_from=home&content_source=1ce6ae8eb85ecab743d40f188e3c0ed6%253ALT5a4a312c5b95dac1e74fb9b2bb56d9fe0afadd6b&logging_key=1ce6ae8eb85ecab743d40f188e3c0ed6%3ALT5a4a312c5b95dac1e74fb9b2bb56d9fe0afadd6b

[warrenaken]

Also an alternative: [https://www.crealitycloud.com/model-detail/birdnet-go-rtsp-microphone-casing-esp32-c6?profileId=6917a720ad9806c8e42f08b8]

[ejwolski]

That looks really good! The one downside I've found to the little case I am using is lack of ventilation - the code does a thermal shutdown at 80C and has to be re-enabled. I may try drilling some holes in the ones I have to allow for some air flow. Cheers!

[truthringr]

That’s the one I ended up printing as well! It’s fairly spacious - I’ve had no issues with temperature even on very hot days - though the antenna may help with that as well.

[ejwolski]

Final (I think) follow up - I replaced the mic on my ESP32C3 (my original wireless mic) with the SHP0645, and the unit is working perfectly like the ESP32C6/SHP0645 combo. Thanks for all the hard work!!!