Audio Data Representation (NCGD009)
1 like198 views
The document discusses audio signal representation and processing. It covers topics like: - Sound is oscillation of atmospheric pressure detectable by human ears within a range of 20-20,000 Hz. - Analog to digital conversion samples an audio signal over time and stores the values. Digital to analog conversion converts these values back to amplitudes. - Common audio data formats include WAV, FLAC for uncompressed time-domain formats, and MP3, Ogg Vorbis for compressed frequency-domain formats. - Multi-channel audio configurations have multiple speakers like stereo, 5.1 surround sound, 7.1, etc. Playback involves a sample pump that converts digital audio to analog for output.
![A low level audio output API
// Main thread
audioAPI_set_callback(&prepare_buffer);
audioAPI_play();
// … some time after (or never):
audioAPI_stop();
// Callback function
static cursor = 0;
void prepare_buffer(
signed_8bit_number *dest_buff,
int length)
{
for (int a=0; a<length; a++) {
cursor ++;
double time = cursor / 48000;
dest_buff[a] = 127*sin(time);
}
}](https://image.slidesharecdn.com/cuncgd009022019-10-15audiodatarepresentation-191016103127/85/Audio-Data-Representation-NCGD009-19-320.jpg)
Audio Data Representation (NCGD009)
- 1. Adam Sporka Charles University Tuesdays 10.40, Room SW1, Malostranské náměstí NCGD009 ’19/’20 Audio Signal Representation
- 2. Technical remark: I drew a number of sketches on the whiteboard. These were not archived in this deck of slides. Sorry about that, I’ll keep it in mind next time!
- 3. Acoustics 101 Sound is oscillation of atmospheric pressure, detectable by ears. 20 Hz – 20,000 Hz
- 4. • Amplitude of anything over time • A/D • Sampling frequency • Bit depth resolution • D/A • Time domain • Frequency domain Signal Processing 101
- 5. • Periodic measurement of the input signal • Storing the values in a sequence A/D Converter 0.000 0.383 0.707 0.924 1.000 0.924 0.707 0.383 0.000 – 0.383 – 0.707 – 0.924 – 1.000 – 0.924 – 0.707 – 0.383 0.000 t
- 6. D/A Converter (DAC) • Processing a sequence of values • Conversion of each value to amplitude 0.000 0.383 0.707 0.924 1.000 0.924 0.707 0.383 0.000 – 0.383 – 0.707 – 0.924 – 1.000 – 0.924 – 0.707 – 0.383 0.000 t
- 7. Audio Data Formats Time domain-based formats • Pulse Code Modulation • WAV • FLAC • ... Frequency domain-based formats • MP3 • Ogg Vorbis • ...
- 8. (Musical Instrument Digital Interface, MIDI) Mentioned for sake of completeness Events and their placement in time
- 9. Multi-channel Audio mono FC stereo FL+FR 2.1 FL+FR+LFE 3.0 FL+FR+FC 3.0(back) FL+FR+BC 4.0 FL+FR+FC+BC quad FL+FR+BL+BR quad(side) FL+FR+SL+SR 3.1 FL+FR+FC+LFE 5.0 FL+FR+FC+BL+BR 5.0(side) FL+FR+FC+SL+SR 4.1 FL+FR+FC+LFE+BC 5.1 FL+FR+FC+LFE+BL+BR 5.1(side) FL+FR+FC+LFE+SL+SR 6.0 FL+FR+FC+BC+SL+SR 6.0(front) FL+FR+FLC+FRC+SL+SR hexagonal FL+FR+FC+BL+BR+BC 6.1 FL+FR+FC+LFE+BC+SL+SR 6.1(back) FL+FR+FC+LFE+BL+BR+BC 6.1(front) FL+FR+LFE+FLC+FRC+SL+SR 7.0 FL+FR+FC+BL+BR+SL+SR 7.0(front) FL+FR+FC+FLC+FRC+SL+SR 7.1 FL+FR+FC+LFE+BL+BR+SL+SR 7.1(wide) FL+FR+FC+LFE+BL+BR+FLC+FRC 7.1(wide-side) FL+FR+FC+LFE+FLC+FRC+SL+SR octagonal FL+FR+FC+BL+BR+BC+SL+SR hexadecagonal FL+FR+FC+BL+BR+BC+SL+SR+TFL+TFC+TFR+TBL+TBC+TBR+WL+WR
- 10. Audio Playback “Sample pump” D/A < audio controller < operating system < process < playback routine
- 13. • Sampling of a signal containing higher frequencies than Nyquist Frequency Aliasing
- 14. Which Sampling Frequency? • Depends on the highest pitch we wish to represent 50 ms 44100 Hz 22050 Hz 14700 Hz 11025 Hz 8820 Hz
- 15. Which Sampling Frequency? 0 s 1 s 2 s 21 KHz 10 Hz
- 16. Which Sampling Frequency? 0 s 1 s 2 s 21 KHz 10 Hz
- 17. Bit Depth • Measured values stored as separated numbers • Resolution in bits – Typical values: 8, 12, 16, 24, 32 bits • Quantization error – measurement error on each sample – signal contains added noise
- 19. A low level audio output API // Main thread audioAPI_set_callback(&prepare_buffer); audioAPI_play(); // … some time after (or never): audioAPI_stop(); // Callback function static cursor = 0; void prepare_buffer( signed_8bit_number *dest_buff, int length) { for (int a=0; a<length; a++) { cursor ++; double time = cursor / 48000; dest_buff[a] = 127*sin(time); } }
- 20. Buffering PLAY BUFFER 1PREPARE BUFFER 2 PLAY BUFFER 2 PLAY BUFFER 3 PLAY BUFFER 4 PLAY BUFFER 5 PREPARE BUFFER 3 PREPARE BUFFER 4 PREPARE BUFFER 5 PREPARE BUFFER 1 PREPARE BUFFER 6 PLAY BUFFER 6PREPARE BUFFER 7 PLAY BUFFER 7 time – dropout CALLBACKS SOUND HW ACTIVITY
- 21. Buffering • Shorter, fewer buffers: – E.g.: 1 buffer, 256 samples @48,000 Hz • 5.33 ms • Almost good for the real-time music performances • Zero margin for errors and delays
- 22. Buffering • Longer, more buffers – E.g.: 10 buffers, 1024 samples each @48000 Hz • 213 ms • Safer • Probably OK for phone calls – Internet radios – 1~60 seconds of buffer • Unsuitable for interactive tasks
- 23. Playback Errors • Data transmission errors – Omission – Pause – Repetition • Wrong sample rate – Different playback speed – Gradual drop of synchronicity
- 24. “Going above Zero” • Distortion: