Add state of the art voice bitrates

[john-peterson]

the current state is one bit per samples. and preferably a cutoff at 8k is recommended to possible unload some work on the encoder. i will try to add opus too. users can record and compae. i am absolutely certain that 16 k is hte right bitrate. hey all we want is les bits right.

i was going to add a volume mixer but i realized how much work it would be since i have t add a pcm recorder and live modify and encode it esentially recreaing the whole class just to fix my mic. i also have a feeling that the right volume can be selected. i have looked at this procedure in my mind; 1 check for voice in the sound. shouldnt be too heavey since google once used that ALL the time remembe hello hgoogle how could it possibly do that without fairies and magic if it did not sample for voice all the time 2 check the amplitute of the voice over a moving average 3 i record long doctors notes and someone has to listen to them and they ften complain that my voice is too low and sometimes too high clipped over modulated aand my phone mic is known as broken mi a2 yeas the famous i dont hear anything i rarely make phone calls but that complain that both recording with screen off and calls are to low.. mostly i notice the recorings because i dont have th it in my mouth i put it down and the wave beam is formed in an... .. etcetc ... thats for another patch now

this was just warmpu. btw im listening to pearlman signal compression to fall asleep. got to chap 4 lz78 compression from anagram dictionaries...

so my big plan that i need is an opus encoder which is more advanceed than amr AND with volume boost that I HAVE NOW so far found anywhere so maybe i will invent the wheel again. theere is one french dev that made it ut he is not responding coincidentally i cycled by him outside in west france two years ago on a trek never made the connection unitl later. .. anyway he has a clean aomplifier for the raw data up to 20 dB and encode it manually in opus... unfortunately he has also made the strange mistake to get stuck in liv and zivs books from 78 possibly amended by welch 79 LZW .... mp3 was t it was mp1 .. anyway it is now much later and opus i much more advanced than amr and ... i need to utilize that power with 16 or 12 kbit encoding.. possibly 16 is enough. he wont let me pass any settings to all the work he done to implement all the encoders in an android project.. its simply 8k cutoff. voce profile. libOPUS has a setting internal settings for voice in addition to manual cutoff. and of course one bit per sample or less. or fe 12 khz/12 kbit and down. i waould use 16/16. still i like one bit per sample. imagine how much heavby lifiting is done to get to that extremey sparse representation .... anyway back to pearlman and.. what els there was something els... right when i change the volume i have o use aLUIT table since a linear volume gain is bad

so dont read any o that drivel above. just tell me WHERE IT IS the stuff i need to do. where is the android audio lib rec lib that can apply a good operational op OP-AMP to the ijmput . the IDEAL op am... nv...

so i am looking for a recorder that applies a volume boost. that will be alot of coding but not difficult. just the tedium o f ordinary work

and secondly i will probably have to create myself that voice colume detector to automatically adjust the gain when i move around in my office and go over all the plates and have the phone stationary on record on the table