How Digital Audio Works - Computerphile

Introduction to Sound and Digital Signal Conversion

In this section, the speaker introduces how sound moves across a room and is converted into an electrical signal that can be used by a computer. The simplest way to convert an analog signal to a digital signal is through the use of a microphone and sound card.

Sound Waves and Electrical Signals

• Air moves in wave patterns across a room, hitting the eardrum and vibrating it into an electrical signal.

• Microphones turn sound into voltage, which can then be converted into ones and zeros for use by computers.

Sample Frequency

• Sample frequency is the number of times per second that the computer will stab in and record levels as numbers.

• Two numbers associated with WAV files are bit depth and sample frequency.

• Sample frequency is one of two parameters used to convert analog signals to digital signals.

• The more slices per second taken by the sample frequency, the better quality audio produced.

Industry Standards

• Two main industry standards for sample frequencies are 44.1 KHz and 48 KHz.

• 44.1 KHz was initially chosen because it was thought to be the highest frequency humans could hear at 22,050 Hz.

• CDs generally come out at 44.1 KHz while DVDs or Blu-rays have different standards.

Conclusion

In this section, the speaker concludes his discussion on sound waves and digital signal conversion.

Final Thoughts

• Use of microphones and sound cards are simple ways to convert analog signals to digital signals for use by computers.

• Sample frequency and bit depth are two important parameters associated with WAV files.

• Industry standards for sample frequencies include 44.1 KHz and 48 KHz, with CDs generally using the former.

• DVDs or Blu-rays have different standards for sample frequencies.

Sampling Rate and Bit Depth

This section discusses the importance of sampling rate and bit depth in digital audio recording.

Sampling Rate

• Microphones can't hear as high as the human ear, so 44.1 or 48 K is well above what you need to pick up sound.

• Sampling at 44.1 or 48 K is sufficient for most audio recordings.

Bit Depth

• Most final products are at 16 bits, which means there are only around 32,000 levels available.

• When dealing with individual instruments, a higher bit depth may be necessary to avoid grainy or distorted sound.

• Professional recordings are generally done at 24 bits, providing over 16 million different levels.

• Recording and mixing at a higher bit depth provides more headroom to work with.

Headroom in Digital Audio Recording

This section explains the concept of headroom in digital audio recording.

Headroom

• Headroom refers to the space between the highest level of a wave and the maximum level that can be recorded without distortion.

• Recording and mixing at a higher bit depth provides more headroom to work with.

Square Waves and Fuzzy Sound

This section discusses how insufficient bit depth can lead to square waves and fuzzy sound in digital audio recording.

Square Waves

• Insufficient bit depth can lead to square waves when points on a wave are connected by lines instead of curves.

• Fuzzy sound can occur at the top end of a fade out if the bit depth is insufficient.

Recording Drums, Guitars, and Vocals

This section explains why recording at a higher bit depth is important when layering different instruments in digital audio recording.

Layering Instruments

• Professional recordings are generally done at 24 bits to provide enough levels for layering different instruments.

• Raw files that haven't been mixed can sound grainy or distorted if recorded at 16 bits.

Digital Clipping

This section discusses the issues of digital clipping and how it can affect audio quality. It also explains how engineers use 24-bit and 48 kHz or possibly 44.1 kHz for most applications.

Digital Clipping

• Digital clipping occurs when you use a digital audio workstation to sum together Wav files constantly.

• Engineers record tracks with enough level that they don't clip but when you sum all these things together, if you've got things that are only a little bit off the top, suddenly you're going to produce a signal which is far in excess of what 24-bit is capable of doing.

• The idea being is some piece of software at the end will show you that it's clipping and you put in a plug-in and turn the whole volume down on the whole thing to get it back so your final output ends up still sitting within the boundaries of what 24-bit is now.

• If you were recording a guitar and all the way through [the] whole thing was just squaring everything off, it might sound pretty nasty.

• You have to make fundamental decisions about what is going to be your sample frequency and what is going to be your bit rate. Generally, for most applications, engineers use 24 bits and 48 kHz or possibly 44.1 kHz.

Passing Information from Phone to Plane

This section discusses how phones can detect magnetic compass heading, pitch angle, roll angle, and acceleration along three axes. The challenge lies in passing this information from phone to plane.

Detecting Magnetic Compass Heading

• Phones can detect magnetic compass heading, pitch angle, roll angle, and acceleration along three axes.

Passing Information from Phone to Plane

• The challenge lies in passing information from the phone to the plane itself.