9

When I drive my kids they often watch cartoons on my android phone with the cartoon's audio coming out of the car speakers, connected to the car's audio system via bluetooth. Unfortunately, there is a very visible 1-2 seconds delay between the video (as observed on the phone's screen) and the audio. There is no such delay when listening to the audio directly from the phone's speaker. There is no noticeable delay when the phone is connected to the same car for the purpose of phone conversations. However, the combination youtube video on the phone's screen + audio from the car's speakers has this annoying 1-2 seconds delay.

Thus the question: what may be causing this and how do I reconfigure whatever needs to be configured in order to fix this?

In case if the specifics matter, the phone is Samsung Galaxy S5, the car is 2013 Honda Odyssey with built-in audio system.

Pᴀᴜʟsᴛᴇʀ2
  • 156,290
  • 29
  • 252
  • 491
Michael
  • 285
  • 2
  • 3
  • 5
  • Have you tried re-pairing the phone to the Odyssey? – Pᴀᴜʟsᴛᴇʀ2 Feb 07 '17 at 21:08
  • I'll remind whoever VTC'd and downvoted, that car accessory questions are on topic for this site. The stereo in the vehicle is an accessory. – Pᴀᴜʟsᴛᴇʀ2 Feb 07 '17 at 22:27
  • 2
    Use headphone jack and aux in put on stereo, see if problem persists. – cory Feb 09 '17 at 18:53
  • None of these answers are helpful. I’ve had this issue with numerous cars. It’s not the phone if video syncs just fine with Bluetooth headsets and some after market car stereos. It seems to just be an individual stereo engineering difference. But what is it? What are some car stereos missing that others and Bluetooth headsets have? – Joe Mielcarek Nov 09 '18 at 15:19

4 Answers4

7

Not wanting to necro a dead thread, but this is a known issue with Honda Bluetooth audio.

There is a bug in the bluetooth handshake that results in data-only transmission. Proof.

This means then that the phone is encoding audio into data, and then the car is decoding that data back into audio. As a result, you get lag on the encode, and lag on the decode. Honda is aware of the issue, but as of 2018 has yet to release a firmware upgrade for cars with this issue. I do not know if 2018 Hondas still exhibit this bug.

MooseLucifer
  • 6,007
  • 4
  • 26
  • 55
lifebinder
  • 86
  • 1
  • 1
2

As far as I know this is just how Bluetooth works. Throughout all my years there has always been a delay in audio when used in combination with Bluetooth.

There are some software that delay the video by a few seconds so it matches up with the audio.

So I guess your fix would be to find a video player where you can manually sync the video and audio.

Maybe VLC on Android can do it: https://play.google.com/store/apps/details?id=org.videolan.vlc

Cucumber
  • 63
  • 1
  • 1
  • 6
  • I have a 2007 Odyssey, and I'm a software developer with 5 years of experience in Bluetooth. A 1 to 2 second delay is not normal. You should expect something in the 100 to 200 millisecond range for Bluetooth. Devices that support A2DP 1.3, however, can compensate for this buffering delay. Both the audio source and sink have to support A2DP 1.3 for it to make a difference. – Samuel Mar 19 '18 at 17:37
1

This is a software / hardware limit problem you could try a better streaming app on your phone if your stereo support others, the problem is streaming audio via Bluetooth is quite a lot of work,

Phone process

Receive audio steam -> convert to a codec for the player -> send over blue tooth

stereo process

receive blue tooth -> convert codec to audio stream -> play audio

so for all sound these 6 steps have to occur and converting audio is a relatively slow process, and made worse by cheap / free codecs.

and more likly to be on your phones end as you say it seems to be bad with you tube so it could be the processing of the video download and then decoding the video codec to display and the processing of the audio for Bluetooth could be that the phone CPU can't cope with that much simultaneously

Barkermn01
  • 295
  • 1
  • 10
  • My iPhone's Music app works just fine with any other Bluetooth speaker, there's no noticeable delay. Also, Bluetooth uses a codec (SBC) that is optimized for minimal CPU usage, not quality. Also see my comment to Cucumber's answer. – Samuel Mar 19 '18 at 17:42
  • Oh your knowing the name of a codec does not change the validity of my answer, the OP said there was more lag when using youtube which means's there is a CPU load issue it does not matter how much it's optimized an ARM CPU is not really built for multitasking high load processes, and just because it's lower quality than MP3 or some other codec does not mean it's not a relatively slow process when compared with input handling., one last thing is not all bluetooth audio uses SBC it is one of a few https://www.soundguys.com/understanding-bluetooth-codecs-15352/ – Barkermn01 Mar 19 '18 at 18:10
  • Any smart phone made, whether Apple or Android, probably since 2010/2011 has more than enough CPU power to handle this use case. In Bluetooth there are 2 profiles for audio transfer, HFP and A2DP. HFP is for phone calls. It is designed for low latency, with a quality tradeoff. A2DP is designed for high quality, with a latency tradeoff. The OP did not make note of the delay with a stand alone music app. A music app, and YouTube, both use A2DP, high latency. A phone call uses HFP, low latency. – Samuel Mar 23 '18 at 15:45
  • 1
    I own an Odyssey. I've used my phone with a lot of other Bluetooth devices. The delay is not because of my phone. The Honda has a superfluously oversized jitter buffer. Perhaps they did this to workaround a hardware issue, e.g. maybe a poor antenna design. Or perhaps the Odyssey's CPU is overloaded and this was the easiest way to overcome audio dropouts. Either way, this is an Odyssey issue. Read a few forums, a lot of people have reported this issue, here's an example: http://www.driveaccord.net/forums/138-audio-electronics-lighting/282537-bluetooth-audio-streaming-a2dp-delay-lag.html. – Samuel Mar 23 '18 at 15:46
  • The OP specifically states Honda Odyssey. Your points aren't wrong, just highly unlikely to be the cause. My point was not to diminish the value of your answer, but instead point blame to the party who's most likely at fault, which in this case is Honda. This is a discussion, not sure why foul language is needed. – Samuel Mar 26 '18 at 16:34
  • Because even there what you have suggested is a breach of Stack Exchange policy no question or answer should be wholly specific. It should have reuse value. This is now the second time of explaining something that is clearly in the policy of Stack Exchange. I don't mind it not being the marked answer what so ever having someone comments with information that will scare away none technical people while making my answer look wholly wrong for anyone experiencing similar issues, comments are to give feedback and discourage use when the answer is wrong. – Barkermn01 Mar 26 '18 at 20:09
  • Please go and read: https://meta.stackexchange.com/questions/184154/closing-changes-on-hold-unclear-too-broad-opinion-based-off-topic-reasons

    I should not have to explain this to user of Stack Exchange Network.

     – Barkermn01 Mar 26 '18 at 20:16
  • For 99% of the people who read this post, your answer will be unuseful. And it is misleading to the non-technical person who doesn't know any better. It's 2018, the age of slow of smart phones has come and gone. – Samuel Mar 27 '18 at 16:40
  • They still on the same drawback as every other ARM Processor is not built to handle large memory altering at speed as it does not use memory directly it goes through stores, the instruction set is limited meaning applications have to use more application space as it's more instructions in memory which in turn means running an application is slower and that's before the application starts reading audio into its memory then transposing it by lots of memory I/O and then writing it to an output buffer then passing it to an OS for Bluetooth transit. and they are asgood as X86 of same Ghz/Cores – Barkermn01 Mar 27 '18 at 18:11
  • Sounds like you know a lot about ARM CPUs. Here's another way to think about it that disproves your incapable phone theory. If the phone had a 2 second lag, then the car would need a 2 second jitter buffer to compensate. If it had a smaller jitter buffer, you would hear audio dropouts. We don't hear that. So it must be that either the car has a 2 second jitter buffer, or both the phone has a 2 second delay and the car has a 2 second jitter buffer. It can't be that only the phone has the 2 second delay, then you would have buffer underruns. – Samuel Apr 19 '18 at 14:13
  • So, in both scenarios the car must add a 2 second delay. It is still possible that the phone has its own 2 second delay, but I've literally tested with hundreds of phones and Bluetooth speakers, this van has the largest delay I've ever seen. Well, on second thought I guess it doesn't disprove your incapable phone theory, but it proves the car is at least part of the reason for the 2 second delay. Maybe your point was that Honda needed the 2 second jitter buffer to compensate for some phone's 2 second delay. – Samuel Apr 19 '18 at 14:22
  • Yeah but the comes back to as you pointed out the car does not accept an audio stream as you pointed out the car is accepting a Raw WAV Byte stream means that 2-second could be buffered out if the app sending the Audio Visual knows about the lag but due to the jailing in Android and iOS it does not app to hardware communication so the app does not know too stagger/delay video by 2 seconds. I know a lot about a lot of CPU architectures and why we have different ones for different processes, in particular, I research and specialise in memory optimisation of applications. – Barkermn01 Apr 21 '18 at 19:49
1

This is a known issue with Hondas. I have much better resource Bluetooth sync in non honda automobiles. The phone isn't the issue it's the honda.

Bill
  • 11
  • 1