audiofiles harika basana (firstname.lastname@example.org ), elizabeth chan (email@example.com ), nikolai...
Post on 19-Jan-2016
Embed Size (px)
AUDIOFILESHarika Basana (firstname.lastname@example.org ), Elizabeth Chan (email@example.com ), Nikolai Sinkov(firstname.lastname@example.org ), Frank Zhang (email@example.com ) 6100 Main Street, Rice University, Houston, Texas 77005
GOALTo explore the MP3 technology and to implement various audio data compressionalgorithms.
Analyze This Audio compression is to compress an audio file into a smaller-sized file.
People cannot differentiate between these two files by just hearing.
Due to its smaller size, the new file can be easily transferred via the Internet.
People try to find better audio compression algorithms that retain satisfying audio quality.
Average Energy Algorithm Zeroes out selected high and low frequencies of the audio file.
Procedure Perform the Discrete Cosine Transform (DCT).
Calculate the signals energy.
Find the mean and the standard deviation of from the energy spectrum.
Keep all frequencies with energies within 1 standard deviation (std) from the mean.
Zero out frequencies with energies outside this range. Similarly, keep frequencies with energies within 2 and 3 stds from the mean.
Perform the Inverse DCT and get the output.
Results Amount of compression is insignificant.
Algorithm would probably work better if the signal is very short, has monotonous tones, and has little noise.
Ding.wav before compressionDing.wav with frequencies within 1 std from the meanDing.wav with frequencies within 2 std from the meanDing.wav with frequencies within 3 std from the mean
Psycho Acoustic Algorithm Linear, tangent or arctangent quantization of the signal.
Procedure Perform the Discrete Cosine Transform (DCT)
Quantize the signal in one of the following ways :
Diagram of the quantization buckets for the three methods
Give certain frequency bands more bits (1000 5100 Hz and 12500 - 15200Hz).
Throw away frequencies below 20Hz and above 20,000Hz. Perform the Inverse DCT.
Results Compression is very significant.
Quality is good for the amount of compression.
Arctangent quantization yields the best quality.
Original signal sampled at 44100Hz The x-axis DT sample and the y-axis is the amplitude After linear quantization After arctangent quantization After tangent quantizationMasking Algorithm The presence of a signal at a particular frequency can raise the perceptual threshold of signals close to the the masking frequency.
Procedure Go through every sample and remove the following samples if they are below a certain threshold.
Results No significant improvement. Need a better way of implementing to get good results.
Conclusion We didnt create MP3 files.
Used the underlying concepts.
Produced much smaller files.
Psycho Acoustic Algorithm is the best, in terms of - amount of compression - sound quality of the output.
Improvements Implement windowing
Implement temporal masking
Bibliography:http://www.sospubs.co.uk/sos/may00/articles/mp3.htmhttp://www.besar.dcs.gla.ac.uk/labs/audiolab/real_site/tutorials/mp3/mp3how.php and more