AUdIoCoUrSeS
Joined: 31 Oct 2002
Posts: 2014
|
| Week 8 - Perceptual coding systems |
 |
|
Perceptual coding systems
Just 13 questions here means elongated answers please, I want you to cover each topic in as much depth as you can, yet concisely.
1. Describe and explain the following perceptual coding systems:
• MPEG, including AAC
• Dolby Digital and AC3
• ATRAC
• Windows Media
• Real Media
• Other systems of current relevance
2. Why is perceptual coding necessary?
3. Describe briefly the use of perceptual coding in the following:
Internet audio
Film sound
DVD-Video
Digital television
Personal stereo / iPOD
4. What can be done, other than perceptual coding, to reduce bitrate?
5. What is masking?
6. How do perceptual coding systems handle signals that are probably going to be masked by other audio?
7. What is Huffman coding?
8. What is the typical bitrate for an MP3 file intended for Internet distribution?
9. What is the bitrate of Dolby AC3 as used in film sound?
10. What is metadata?
11. Explain the principles of predictive coding.
12. Explain the basic difference between downloading text & graphics files as compared to the streaming of a sound file over the internet.
13. Describe the process of creating a DVD master.
_________________
It's all in the ears. - Learn the concepts not the software.
Audio Courses is a way into the music business for you
|
Mon Oct 25, 2004 11:19 am |
|
|
iNSTiNCT2765
Joined: 05 Nov 2003
Posts: 60
Location: Denmark |
|
|
|
Week 8 – Perceptual Coding Systems
1. Describe and explain the following perceptual coding systems:
• MPEG, including AAC
The Motion Picture Experts Group introduced the MPEG codec in 1988. There are three standards:
MPEG-1 (coding of mono and stereo signals at sampling rates of 32, 44.1 and 48 kHz)
MPEG-2 (ISO/MPEG IS-11172: coding of 5+1 multi-channel sound signals and low bit rate coding of mono and stereo audio at sampling rates of 16, 22.05 and 24 kHz)
MPEG-4 (ISO/IEC 14496)
Coding methods in MPEG include Layer3 (MP3), which attains high compression ratios while preserving audio quality by using psychoacoustic masking, redundancy reduction, bit reservoir buffering and joint-stereo.
AAC (MPEG4) is another coding method introduced in 1997. It is an international standard and uses temporal noise shaping, backward adaptive linear prediction and enhanced joint stereo coding techniques. AAC supports a wide range of sampling rates (8–96 kHz), bit rates (16–576 kbps) and from one to 48 audio channels. AAC has 100% more coding power than Layer 2 and 30% more coding power than Layer 3 (MP3).
• Dolby Digital and AC3
AC3 is the third generation audio coding algorithm from Dolby. The audio spectrum is divided into narrow frequency bands optimised with the respect to the frequency selectivity of human hearing. This means this codec can sharply filter coding noise and the quality of the audio can be preserved.
• ATRAC
Adaptive Transform Acoustic Coding is Sony’s coding method for Minidisc’s. Audio is compressed to 1/5 the size of 44.1KHz 16 bit linear audio with virtually no loss in the sound quality.
• Windows Media
Windows Media Audio (WMA) is a codec developed by Microsoft used in conjunction with Windows Media Player. There is a 20:1 data compression using this codec and this makes this type perfect so streaming audio for auditioning purposes (audio clips of a CD for preview before purchasing). Some MP3 devices can also read this codec so it offers an alternative to encoding using MP3.
• Real Media
The RealNetworks are the originators of streaming audio over the Internet. The Real Audio codec is used most commonly for this purpose and is rarely used for CD-ripping or for storage of media.
• Other systems of current relevance
The Ogg Vorbis codec is a free open source codec and can compete with MP3 and WMA codecs. Many third party software such as Sound Forge support the format.
2. Why is perceptual coding necessary?
Perceptual coding is necessary to reduce the data rate of media without any noticeable loss in the quality of it. Perceptual coding takes advantage of the phenomenon of auditory masking.
3. Describe briefly the use of perceptual coding in the following:
Internet Audio – Audio streamed over the Internet needs to be coded in order for the data to be transferred seamlessly so there are no interruptions in the audio. The audio therefore needs to be reduced in data rate but needs to be of a high enough quality so no obvious loss is noticeable. This is why codecs, which use perceptual coding technology, are used for such files. Codecs used in audio files that are streamed over the Internet include MP3, Real Audio, WMA and Ogg Vorbis.
Film sound – Dolby AC3 codec is used for the surround sound of films. This offers high quality surround sound at a very low data rate.
DVD-Video – This is encoded using MPEG2 Variable Bit Rate Encoder. This encodes the video retaining quality but reducing data by discarding video data that is repeated. The reason for the variable bit rate encoding is that some scenes are more complex than others and require more bits while simple scenes don’t need as many bits and reducing the bits saves space on the disc.
Digital television – Dolby AC3 is also used for Digital TV but since it has a limited bandwidth, the bit rate is somewhat lower than that of DVD.
Personal stereo / iPOD – Personal stereo’s such as a Minidisc, MP3 or Apple iPOD player all use codecs to reduce the data size of the audio with no or little noticeable loss in quality. The Minidisc uses ATRAC coding that reduces the size of a 16 bit, 44.1 KHz linear audio file to 1/5 its original size. MP3 players use the MPEG Layer 3 (MP3) codec for a high data compression with no loss in the quality. Some MP3 Players can also read audio files coded using the WMA codec. The iPOD can play files coded using AAC.
4. What can be done, other than perceptual coding, to reduce bitrate?
Predictive coding can also be used to reduce bitrate. It is a data reduction method, which has a ‘predictor’ that has knowledge of typical audio signal behaviour. The ‘predictor’ looks at the preceding audio signal and attempts to determine what will happen next and due to the repetitive nature of audio signals, the prediction is usually very accurate. The difference between the original signal and the prediction is then recorded as the data reduced signal. To regenerate the signal, a decoder must use the same ‘predictor’ knowledge and this is usually done to an accuracy of 98%. This accuracy is completely dependant on the predictor algorithm. To improve the precision of the system, a multi-band coding technique is also being used. This means splitting the signal into four frequency bands and applying separate predictors for each. One thing to be taken into consideration is that the encoder and decoder of the system must use the exact same predictor. Therefore, both have to be updated in order to ensure no signal degradation. The predictive coding system provides a 4:1 data reduction.
5. What is masking?
When the ear hears two tones with the same dominant frequencies, the louder tone will make the softer tone less sensitive to the ear and it will be masked so only the louder tone is audible. Lossy data compression methods, such as perceptual coding, take advantage of this phenomenon, since the human ear does not hear the masked tones anyways, they don’t need to be coded, therefore saving storage space.
6. How do perceptual coding systems handle signals that are probably going to be masked by other audio?
Using algorithms, the perceptual coding system determines the masked signal. Since the system knows that a masked signal won’t be audible to humans, it leaves it out of the coding to reduce the data.
7. What is Huffman coding?
Huffman coding is a method of coding introduced by David Huffman in a seminal paper in 1952. It works by varying the length of the encoded symbol in proportion to its information content. MP3 encoding takes advantage of this method.
8. What is the typical bitrate for an MP3 file intended for Internet distribution?
128Kbs
9. What is the bitrate of Dolby AC3 as used in film sound?
In cinemas, Dolby AC3 uses a 640Kbs bitrate. On DVD’s and laserdiscs, the bitrate is 384Kbs.
10. What is metadata?
Metadata is data that comes with the audio. It holds information about the work such as authoring information (artist, label, engineer, year of release, genre etc.). Metadata can also be used to control certain characteristics of the audio. Metadata is documentation used for cataloguing the audio for preservation reasons.
11. Explain the principles of predictive coding.
Explained in question 4.
12. Explain the basic difference between downloading text & graphics files as compared to the streaming of a sound file over the Internet.
The basic difference here is that when downloading text or graphic files, the files are downloaded completely to the computer before you can use them as where a streaming sound file is played as it is downloading.
13. Describe the process of creating a DVD master.
Creating a DVD master is more complex than just creating a CD master. The process includes:
Audio And Video Preparation/Encoding:
The audio and video data is encoded using a MPEG2 compression method. This reduces the data rate and saves space on the disc without any loss in the quality of the media. Video is encoded using the MPEG2 Variable Bit Rate Video Encoder, which reduces the video data while obtaining the maximum quality by leaving out redundant visual information and giving more bits for complex scenes and less for simple scenes. The MPEG2 Variable Bit Rate Audio Encoder is used to encode multichannel audio (5.1 or 7.1).
Subtitling/Disc Definition:
All included language versions of the subtitles are formatted in the Disc Definition stage with the other basic information about the disc.
Multiplexing:
The multiplexer takes all the audio/video and subtitle data and combines it into one single digital data stream.
Disc Building:
The encoded and multiplexed data is used by disc building software to create a final disc image. Here, information such as file structuring, synchronization of time codes, disc branching details for interactivity and disc content specified at the Disc Definition stage are combined.
Simulation:
Before the disc is pressed in volumes for manufacturing, a computerised simulation of the disc playback is carried out to confirm the data on the disc is correct.
Channel Encoding:
After the simulation has verified that the data on the disc is correct and the disc image has been created, error correction data must be added to the data stream. This is called channel encoding and it writes the DVD formatted data stream to the Laser Beam Recorder, which is used to cut the glass master DVD.
Last edited by iNSTiNCT2765 on Sun Nov 07, 2004 5:26 am; edited 1 time in total |
|
Sun Oct 31, 2004 9:36 am |
|
|
|
|
AUdIoCoUrSeS
Joined: 31 Oct 2002
Posts: 2014
|
| DVD audio |
|
|
OK it seems you need some help with DVD audio right?
Check this site out, it has further links also.
It is an area to begin looking at as a professional.
http://www.doom9.org/ the forums are very helpful there.
I imagine there will be DVD questions in the exam!
_________________
It's all in the ears. - Learn the concepts not the software.
Audio Courses is a way into the music business for you
|
|
Mon Nov 01, 2004 11:04 am |
|
|
iNSTiNCT2765
Joined: 05 Nov 2003
Posts: 60
Location: Denmark |
There, the answers are now all complete!  |
|
Sun Nov 07, 2004 5:27 am |
|
|
|
|
|
|
All times are GMT. The time now is Fri May 16, 2008 5:17 pm
|
|
|
|
| |
