Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Which Encoder Has The Best Block Switching? (Read 2110 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Which Encoder Has The Best Block Switching?

I ask this because I'm thinking of adapting a perceptual encoder for use as the front end of a system for computer simulation / implementation of human psychoacoustic grouping rules I'm trying to make for the final year project. What I'd want to do is to take the output from its filterbanks for use in the next stage (before the signal goes through psychoacoustic compression algorithms). A filterbank with block switching would give the system preliminary information about where the attack of an instrument is located and also give better time resolution to these bits for analysis.

Any idea whether this is possible at all? (Altering the code so that it outputs the FFT-ed waveform (er, or something more like a spectrogram at this stage) with short / long block assignments without already compressing. I get the feeling that block lengths are chosen in the middle of the psychoacoustic compression process? Also, I guess I'm limited to encoders whose source code is open.)

Which Encoder Has The Best Block Switching?

Reply #1 –
Quote
I get the feeling that block lengths are chosen in the middle of the psychoacoustic compression process? Also, I guess I'm limited to encoders whose source code is open.)

The block length may not be necessary be chosen from the psychoacoustic model.
If you look at the block diagram of MP4 AAC SSR, the block switching decision is taken from the subband filterbanks instead. In fact, I think by splitting the signal into 4 subbands, would increase the transients  isolation from the huge low frequency components.. Certain percussion instruments may be hidden by the huge low frequency components. 

I think there is still some limitation to the ISO based PE calculation for block switching decision.

wkw

Which Encoder Has The Best Block Switching?

Reply #2 –
where can I find some information about the rules by which psychoacoustic encoders make decisions about block switching?

Based on your comments on my other thread it seems mp3 has potential for the best block switching, if only it were not limited in other ways... perhaps one of the modified variants of mp3?

Of course the state of development of the actual block switching algorithm in the encoder is as important as the format.

Which Encoder Has The Best Block Switching?

Reply #3 –
I still think the switching strategy may depend on the encoder format itself...
I still don't like switching to short-block as this will limit the bitrates. Short-blocks consumed too much bits compared to long block.. Also, it has poor tone isolation.

I think block switching decisions are implementation dependant.. Different encoders may implement it differently..

wkw