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: LC or LTP (Read 3926 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

LC or LTP

So which one is better? Or better question, how does Longterm Prediction affect the quality?

LC or LTP

Reply #1 –
Quote
Originally posted by Tom Servo
So which one is better? Or better question, how does Longterm Prediction affect the quality?


LTP has better quality.

But LC is playable with hardware players and MPEG4 video players. (The only MPEG4 player that supports LTP is MPEG4ip, since it uses FAAD for decoding)

You should try them for yourself. I, personally, can't hear big quality differences between LTP and LC.

Regards;

Roberto.

LC or LTP

Reply #2 –
I would recommend using LC mode, because of compatibility issues with the hardware players and other software AAC decoders.

LTP and other prediction tools are usable at very low bitrates - not for VBR or medium / high bitrates.

LC or LTP

Reply #3 –
Well, I'm encoding for local storage only and not to play them back on hardware player (I don't intend to buy one). At bitrates like 144kbit or 160kbit is there no improvement at all?

LC or LTP

Reply #4 –
The LTP tools actually improves coding of audio clips that contain lots of periodic
information such as human speech which contain pitch information. If you code
a clip of human speech using MP3 at very low bitrate, eg: 8kbps mono and compare with standard speech algorithm such as the CELP ITU_T G723 at 5.6kbps mono, you will notice that the MP3 coded clip sounded as if the person is speaking with "blocked- nose". It sounded dull or flat. Most speech compression algorithm incorporates a LTP in their designs. Certain musical clips that contain human voices will definately benefit from the LTP.

Why couldn't the psychoacoutics model compensate for pitch information ? That I do not know.  Maybe it has something to do with the modelling approach. LTP actually models the human vocal tract whereas psychoacoustics models is based on human hearing.


KWW

LC or LTP

Reply #5 –
Well, somehow it seems LTP doesn't actually work on Aacenc, I checked it and it seems that only a tag in the frame header is changed?