Audio Share can share Windows/Linux computer's audio to Android phone over network, so your phone becomes the speaker of computer. (You needn't buy a new speaker😄.)
Apache License 2.0
847
stars
40
forks
source link
[New Feature] Add ultra low latency mode for using the phone as a substitute speaker during gaming #78
Describe the new feature you'd like
I propose to add a ultra low latency mode for example first person shooter gaming
Describe the reason why you want to add the feature
I need this feature because if we want to play games using phone as the speaker with the help of this software, we need to reduce latency as much as possible even at the cost of audio quality and increased data traffic between devices
Describe how to implement the feature if you can
we can use 16bit 22050Hz audio as a base input format (maybe use simple dithering algorithm to downsample original audio)
then use some very low complexity codec (such as ADPCM) to transmit the audio data to the phone
then allow user to finetune the exact AudioTrack buffer size or even use alternative libraries like OpenSL ES to reduce latency on decoding or use software decoding to bypass hardware audio decoder that potentially has more latency
finally use tricks to bind the audio process to the largest CPU core, for example, Cortex A78 instead of Cortex A55
Additional context
16bit 22050Hz sounds good enough for even some very modern shooters like Cyberpunk 2077 when latency is a priority
if we need to further reduce latency, we can use even more aggressive codecs like AMR-WB
mkckr0
commented
1 month ago
Describe the new feature you'd like I propose to add a ultra low latency mode for example first person shooter gaming
Describe the reason why you want to add the feature I need this feature because if we want to play games using phone as the speaker with the help of this software, we need to reduce latency as much as possible even at the cost of audio quality and increased data traffic between devices
Describe how to implement the feature if you can we can use 16bit 22050Hz audio as a base input format (maybe use simple dithering algorithm to downsample original audio) then use some very low complexity codec (such as ADPCM) to transmit the audio data to the phone then allow user to finetune the exact AudioTrack buffer size or even use alternative libraries like OpenSL ES to reduce latency on decoding or use software decoding to bypass hardware audio decoder that potentially has more latency finally use tricks to bind the audio process to the largest CPU core, for example, Cortex A78 instead of Cortex A55
Additional context 16bit 22050Hz sounds good enough for even some very modern shooters like Cyberpunk 2077 when latency is a priority if we need to further reduce latency, we can use even more aggressive codecs like AMR-WB