Technology - Audio-Visual Coding 1 - MPEG 1 and MP3

Synopsis

We discuss MPEG-1 and MPEG-2 audio compression (coding) and major non-MPEG audio coding systems including Real Audio, Windows Media and Vorbis. A detailed explanation of the internal mechanism of MP3 (MPEG-1 Audio Layer III) compression serves as a basis for understanding MP3pro and all other perceptual audio codecs, including Advanced Audio Coding (AAC).

Table of Contents

• 1. Synopsis
• 2. Introduction
  • 2.1 Open standards and licensing
• 3. Data compression and ‘coding’
  • 3.1 64kb/s voice with companded 8 bit samples
    • 3.1.1 Simple ADCPM compression for voice
  • 3.2 Sample rates and word sizes for music
    • 3.2.1 24 bit samples at 96kHz
    • 3.2.2 Limits of lossless audio compression
• 4. MPEG-1
• 5. MPEG-1 Audio Layers I and II
• 6. MPEG-I Audio Layer III (MP3)
  • 6.1 Stereo correlations
    • 6.2 Temporal masking
    • 6.3 Frequency masking
    • 6.4 Hearing threshold
    • 6.5 Time to frequency domain conversion
    • 6.6 Frames and granules
    • 6.7 Polyphase filter bank
    • 6.8 Modified Discrete Cosine Transform (MDCT)
    • 6.9 Normal and short windowing
    • 6.10 Omission of inaudible frequency terms
    • 6.11 Quantisation of audible frequency terms
    • 6.12 Optimising the trade-offs
    • 6.13 Variation in encoders and decoders
    • 6.14 The final frame of data
    • 6.15 Variable Bit Rate (VBR)
    • 6.16 The decoder
    • 6.17 mp3PRO
    • 7. Other lossy audio compression standards
      • 7.1 Real Audio
      • 7.2 Windows Media
      • 7.3 Ogg Vorbis
    • 8. Related reports