Hello, I am Jose Garcia, an electrical engineering student at Rice University in Houston, TX. I am doing some research which involves decoding and encoding MP3 files. I learned about the LAME encoder recently, and it seems like a great tool. However, I have some questions about the inner works of LAME which I could not resolve simply online. I would greatly appreciate answers to the following questions.
1. What are the basic step by step functions that LAME uses for encoding WAV files into MP3's? For example, first sampling, then quantization, application of filters, etc. 2. Since MP3 uses forward adaptive bit allocation, encoders quantize and perform all calculations, and the bit allocations are encoded and sent to the decoder. How exactly does the ?encoding? actually happen in LAME? For example, do you use some sort of Huffman coding? If it is not a well known variation, what are the properties and basic formula? 3. What transform does LAME use? (such as MDCT) 4. What digital filters does LAME use, and for what purpose? (low-pass, band-pass, etc.) 5. How is subband coding used in LAME? 6. How does GPSYCHO use psycho-acoustics? For example, does it have a set table of values which it compares audio to? If so, where can I find information about the table of values used? 7. Is the high-pass filter for GPSYCHO implemented yet, or is it still just a low-pass filter? If a high-pass filter is implemented, what version of LAME includes the additional filter? 8. What is the basic formula for the GPSYCHO filters? I apologize if this is the wrong email to send these questions. If so, I would appreciate it if you forwarded this email to anyone who can answer the questions or if I am sent an alternate email address. Thanks, Jose _______________________________________________ mp3encoder mailing list mp3encoder@minnie.tuhs.org https://minnie.tuhs.org/mailman/listinfo/mp3encoder