Audio Fundamentals: Waves, Sampling, Formats & Applications

Core Concepts: Sound as a Physical Wave

How Sound Travels

Sound is a longitudinal mechanical wave that propagates through a medium (air, water, or solids) by compressing and rarefying particles. Key properties:

• Frequency (f): Measured in Hertz (Hz), determines pitch (e.g., 261.63 Hz = middle C).
• Amplitude (A): Measured in Pascals (Pa), determines loudness (decibels, dB).
• Wavelength (λ): Distance between two identical points in a wave cycle (λ = v/f, where v = speed of sound ≈ 343 m/s in air at 20°C).

graph LR
    A["Sound Source\n(e.g., vocal cords)"] -->|"Vibration"| B["Medium\n(air/water/solid)"]
    B -->|"Compression/Rarefaction"| C["Wave Propagation"]
    C -->|"Detected by"| D["Human Ear\nor Microphone"]

Shows compression/rarefaction cycles, wavelength, amplitude, and frequency markers. (Image: Kulayada, CC BY-SA 4.0, via Wikimedia Commons)

Worked Example: Ncell’s Ringtone Pitch

Ncell’s default ringtone uses a 440 Hz sine wave (A4 note). If the speed of sound is 343 m/s:

• Wavelength (λ) = v/f = 343/440 ≈ 0.78 m (78 cm).
• A shorter wavelength (e.g., 176 Hz) would produce a deeper bass note (C3).

Digital Audio: From Analog to Binary

Sampling and Quantization

1. Sampling: Converting continuous analog signals into discrete values at fixed intervals (sample rate, e.g., 44.1 kHz = 44,100 samples/sec).
2. Quantization: Assigning each sample a discrete value (bit depth, e.g., 16-bit = 65,536 possible levels).

Nyquist Theorem: To avoid aliasing, the sample rate must be at least twice the highest frequency in the signal.

• Example: For human hearing (20 Hz–20 kHz), minimum sample rate = 40 kHz.

graph TD
    A["Analog Signal\n(smooth wave)"] -->|"Sampling"| B["Discrete Samples\n(44.1kHz)"]
    B -->|"Quantization"| C["Digital Values\n(16-bit)"]
    C -->|"Binary Encoding"| D["MP3/WAV File"]

Bit Depth and Dynamic Range

Worked Example: WhatsApp Voice Notes

• WhatsApp compresses voice notes to 16 kHz sample rate, 16-bit depth (mono).
• File size saved: Lower sample rate reduces data by 50% vs. CD-quality (44.1 kHz).

Audio File Formats: Trade-offs in Compression

Uncompressed Formats (Lossless)

Compressed Formats (Lossy)

How MP3 Works:

1. Frequency Division: Splits audio into 32 bands.
2. Perceptual Noise: Removes frequencies masked by louder sounds (e.g., bass drowning out highs).
3. Bit Allocation: Allocates bits to perceptually important bands.

Worked Example: NEPSE Stock Alerts

• NEPSE’s audio alerts use MP3 at 64 kbps to balance clarity and file size.
• Original: 16-bit WAV (1.4 MB/min).
• Compressed: MP3 (96 KB/min) → 93% smaller.

Audio Hardware: Microphones and Speakers

Microphone Types

Speaker Systems

Worked Example: Daraz’s Call Center

• Uses full-duplex audio (separate mics/speakers) for simultaneous customer-agent communication.
• Latency < 50 ms ensures natural conversation flow.

Audio in Multimedia Applications

Real-World Applications

1. eSewa Voice Authentication:

   • Uses formant analysis (vocal tract shapes) to verify identity via voiceprints.
   • Sample rate: 8 kHz (sufficient for speech recognition).

2. Pathao Driver Navigation:

   • Directional audio cues (e.g., "Turn left in 100m") via bone conduction headphones.
   • Compression: AAC at 128 kbps for real-time clarity.

3. NTC’s Emergency Alerts:

   • Dual-tone multi-frequency (DTMF) for automated calls (e.g., "Your bill is due").
   • Example: *123# generates tones for 1 (1209 Hz) and 2 (1336 Hz).

Audio Editing Tools

Exam Tip

1. Define Key Terms:

   • Aliasing: Artifacts from insufficient sampling (e.g., a 1 kHz sine wave sampled at 1.5 kHz appears as 500 Hz).
   • Psychoacoustics: How humans perceive sound (used in MP3 compression).

2. Calculate File Sizes:

   • Formula: File Size (bytes) = (Sample Rate × Bit Depth × Channels × Duration) / 8
   • Example: 30 sec stereo WAV at 44.1 kHz, 16-bit: = (44100 × 16 × 2 × 30) / 8 = 1,058,400 bytes ≈ 1 MB.

3. Compare Formats:

   • MP3 vs. WAV: MP3 is lossy (smaller file) but loses quality; WAV is lossless (larger file, pristine).
   • AAC vs. OGG: AAC is more efficient for streaming (used by Apple), OGG is open-source.

4. Real-World Scenarios:

   • Always relate to Nepali apps (e.g., "Why does Khalti use AAC for voice prompts?" → Lower bandwidth).
   • Diagrams: Draw wave forms, polar patterns, or speaker setups in exams.

Visual Summary:

mindmap
  root((Audio Fundamentals))
    Physics
      Wave Properties
      Nyquist Theorem
    Digital Conversion
      Sampling
      Quantization
    Formats
      WAV
      MP3
      AAC
    Hardware
      Microphones
      Speakers
    Applications
      eSewa Voice
      Pathao Audio
      NTC Alerts