Technology
Technology
Technology
Technology
Digital Audio: File Formats
Digital Audio: File Formats
Digital Audio: File Formats
Digital Audio: File Formats
Introduction
From the introduction of recorded audio in the early 1900s until the 1980s, everything was analog. The waveform was captured as a groove on vinyl or a magnetic signal on magnetic tape. But in 1982, that all changed with the introduction of the CD. Technological advances in digital music continued and today almost all consumer music today is recorded digitally. Digital music is available in various formats. Different formats offer different levels of quality, file size, and ease of encoding and decoding. When choosing the best format for listening, it’s a good idea to understand the basics of digital recording and the differences between the common audio file formats.
What is Sample Rate? What is Bit Depth? - Digital becomes Analog
With analog records, there is a direct waveform shape represented by a vinyl record's groove or the magnetization amount of a cassette tape. Analog is fantastic, but it does struggle with the issue of noise artifacts. When analog recording media is copied, more noise is introduced, and the overall fidelity is decreased. That’s why most records are created from a single master record.
But digital audio recording uses something called digital sampling. This reduces a continuous waveform signal into a discrete-time signal. A waveform is sampled at discrete and evenly-spaced intervals along the waveform, known as the sample rate. For example, a sample rate of 44.1 kHz captures 44,100 samples per second. 44.1 kHz is the industry standard for consumer records, but 48 kHz is also popular. Higher sample rates are used for recording (or audiophile projects), but later downsampled to make consumer CDs and other digital formats.
Each sample is given a precise number that represents the value of the signal at that point. The accuracy of this sample value is called the bit depth. For example, a 16-bit bit depth represents the sample value with a 16 digit binary (1s and 0s) number. 16-bit recording offers 65,535 audio amplitudes while 24-bit recording offers over 16 million values. The 32-bit float method is the latest and greatest way to store sample values. This method can be a bit confusing, but there are a few important things to note:
Float-point format – sample values stored with scientific notation, meaning they can be represented as very small and very large numbers. Much better than the fixed-point method.
Dynamic range – the possible dynamic range is increased to 1528 dB, up from 96 dB. The loudest sound possible is roughly 194-210 dB. This means the 32-bit method can support any and all sounds.
Large file sizes – using the 32-bit float for audio files means the files will be roughly 50% larger than before (with a 48 kHz sampling rate).
Avoid clipping or distortion – significantly more headroom means there’s no need to worry about the gain to avoid clipping and other distortions.
Digital records have the obvious benefit of being stored on digital storage formats like CDs or digital audio tapes (DAT). Digital records can also be used to listen on various electronic devices, including mp3 players, smartphones, and computers. But digital recordings have another benefit. They can be copied and mixed without introducing additional noise to the recording.
What is an Audio Format?
An audio format is a specific way to store audio. Different formats have different ‘strengths’ and ‘weaknesses’ in that they are better or worse at storing information or easier or harder to unpack. An audio format can be defined by its quality and fidelity.
The audio reproduction chain isn’t perfect, but it’s improved in the last 20 years to produce better results and higher fidelity recordings. Today there are many audio formats, but they can be split into three groups based on fidelity and amount of compression.
Uncompressed Lossless Formats
Lossless audio files that have not been compressed in any way are the best audio format. None of the audio information has been lost to a compression algorithm. The most common lossless PCM (pulse-code modulation) audio formats are: WAV, AIFF, and AU.
Raw Headerless PCM
Raw headerless PCM is a digital audio that still exists in its raw pulse-code modulation form. This format has no metadata header information meaning there is no sample rate, bit depth, number of changes, or other information stored with the raw audio. If you want to listen to this format, you need to provide the header information for playback. This format is as unaltered as it gets and it’s how sound is captured for DVDs and CDs.
WAV
The WAV (Waveform Audio File Format) format is a lossless audio format. WAV files are essentially an audio wrapper, or container, for audio. It’s very popular today and is best when you need to retain the highest audio quality. Uncompressed WAV files are encoded with pulse code modulation, which is just sampling along an analog recording at regular intervals (referenced above).
AIFF
AIFF is the Mac counterpart to WAV. AIFF, which stands for Audio Interchange File Format, was developed in the late 1980s by Apple. Like WAV, it’s a wrapper that can hold multiple kinds of information, but is usually for uncompressed PCM audio.
Compressed Lossless Formats
Digital files can be compressed using a lossless algorithm to create a lossless format like FLAC, WMA, or ALAC. Lossless does not lose any information during compression because it shrinks only the file size, not the content of the audio file. This compression process works much like a ZIP or RAR file. Lossless formats retain the peaks and lows, but have much smaller file sizes for easy sharing and storage.
FLAC
The FLAC (Free Lossless Audio Codec) format is extremely popular because it can compress the original file by up to 50% without compromising the integrity of the audio. It’s also open-source and supports a range of sampling rates and bit depths that keep audio enthusiasts happy.
ALAC
ALAC is the lossless audio compression format created by Apple, hence the name Apple Lossless Audio Codec. All audio in the Apple Music catalog is encoded using ALAC. It’s comparable to FLAC although FLAC is slightly easier to unpack and playback. ALAC supports bit depths up to 32-bit and has a maximum sample rate of 384 kHz (384,000 samples per second!).
Compressed Lossy Formats
Lossy compression, on the other hand, removes information from the audio file itself. The peaks and lows are removed or clipped. Anything that is deemed unnecessary because it is at the fringes of human hearing. Lossy formats include MP3, OGG, and AAC.
MP3
MP3 stands for MPEG-1 Audio Layer-3. It is the most common compression format when the audio file needs to be very small. It was popular in the early 2000s but a trend towards higher-fidelity records has pushed it out of favor. MP3s are 1/12th the size of the original format but important audio information is lost during compression. The MP3 format supports sample rates from 16 to 48 kHz.
OGG
The OGG format, also known as Ogg Vorbis Compressed Audio File, is an open-source audio container format. The audio is saved in the Ogg container and compressed with the Vorbis audio compression algorithm. It’s very comparable in quality to MP3s.
AAC
The Advanced Audio Coding (AAC) format is essentially the successor to the MP3. It stores audio information in a similar sized file, with the .m4p or .m4a file extension, while retaining a higher sound quality. It can support up to 96 kHz sample rates and up to 48 channels.
WMA
WMA stands for Windows Media Audio. WMA is an audio file wrapper and audio codec. It’s used by the Windows Media Player that is built into all Windows operating systems. WMA is a lossy format, but there is a lossless version called Lossless WMA.
Lossy compression reigned supreme in the early days of digital music when the storage space available on hard drives was much less than it is today. While MP3s are better from a storage perspective, they are not ideal for listening. The lossy compression method introduces artefacts that create a sizzling sound. MP3s are also woefully flat sounding. Lossy formats don’t produce the best listening experiences and can leave listeners feeling worse than before listening. Lossless compression should be used wherever possible.
The Audio Reproduction Chain - The Sound Machine
Consumers listen to all modern music after it’s passed through the audio reproduction chain. New music is recorded using analog instruments and amplifiers that is immediately turned into digital information. Existing music is remixed or reproduced using a master record, either a master vinyl for older records or a master 32-bit 192 kHz digital recording.
From there, digital recordings are usually interpolated and downsampled and possibly compressed into more consumer-friendly formats. This is also when mixing and other alterations to the raw audio file are generally added. Most listeners use low-grade playback devices like headphones, earbuds, or simple stereo speakers that won’t reveal the benefits of a 32-bit 192 kHz recording. Instead, the industry standard is 44.1 kHz or 48 kHz, which is found on CD recordings and MP3s. Millions of albums have been released in the 44.1 kHz format.
When a digital file is finally in the hands of a consumer through their music library or a streaming service, it is converted back to an analog waveform using a DAC. This transforms the digital information, stored as 1s and 0s, into a voltage or current modulated electrical signal. The waveform is then transmitted through the amplifiers of headphones or speakers for listening.
There are various audio formats acting as different storage containers for digital audio information. Different formats have different ‘strengths’ and ‘weaknesses’ in that they are better or worse at storing information or easier or harder to unpack.
It’s difficult for the consumer to differentiate these differences after passing through the audio reproduction chain. But as a general rule, most of the issues or unwanted artefacts of the reproduction chain are added when downsampling, compressing or mixing digital audio. Ideally, audio files should be left as raw as possible to preserve the original characteristics. This means lossless audio formats are preferred and compressed lossy formats should be avoided for the best listening experiences.
Introduction
From the introduction of recorded audio in the early 1900s until the 1980s, everything was analog. The waveform was captured as a groove on vinyl or a magnetic signal on magnetic tape. But in 1982, that all changed with the introduction of the CD. Technological advances in digital music continued and today almost all consumer music today is recorded digitally. Digital music is available in various formats. Different formats offer different levels of quality, file size, and ease of encoding and decoding. When choosing the best format for listening, it’s a good idea to understand the basics of digital recording and the differences between the common audio file formats.
What is Sample Rate? What is Bit Depth? - Digital becomes Analog
With analog records, there is a direct waveform shape represented by a vinyl record's groove or the magnetization amount of a cassette tape. Analog is fantastic, but it does struggle with the issue of noise artifacts. When analog recording media is copied, more noise is introduced, and the overall fidelity is decreased. That’s why most records are created from a single master record.
But digital audio recording uses something called digital sampling. This reduces a continuous waveform signal into a discrete-time signal. A waveform is sampled at discrete and evenly-spaced intervals along the waveform, known as the sample rate. For example, a sample rate of 44.1 kHz captures 44,100 samples per second. 44.1 kHz is the industry standard for consumer records, but 48 kHz is also popular. Higher sample rates are used for recording (or audiophile projects), but later downsampled to make consumer CDs and other digital formats.
Each sample is given a precise number that represents the value of the signal at that point. The accuracy of this sample value is called the bit depth. For example, a 16-bit bit depth represents the sample value with a 16 digit binary (1s and 0s) number. 16-bit recording offers 65,535 audio amplitudes while 24-bit recording offers over 16 million values. The 32-bit float method is the latest and greatest way to store sample values. This method can be a bit confusing, but there are a few important things to note:
Float-point format – sample values stored with scientific notation, meaning they can be represented as very small and very large numbers. Much better than the fixed-point method.
Dynamic range – the possible dynamic range is increased to 1528 dB, up from 96 dB. The loudest sound possible is roughly 194-210 dB. This means the 32-bit method can support any and all sounds.
Large file sizes – using the 32-bit float for audio files means the files will be roughly 50% larger than before (with a 48 kHz sampling rate).
Avoid clipping or distortion – significantly more headroom means there’s no need to worry about the gain to avoid clipping and other distortions.
Digital records have the obvious benefit of being stored on digital storage formats like CDs or digital audio tapes (DAT). Digital records can also be used to listen on various electronic devices, including mp3 players, smartphones, and computers. But digital recordings have another benefit. They can be copied and mixed without introducing additional noise to the recording.
What is an Audio Format?
An audio format is a specific way to store audio. Different formats have different ‘strengths’ and ‘weaknesses’ in that they are better or worse at storing information or easier or harder to unpack. An audio format can be defined by its quality and fidelity.
The audio reproduction chain isn’t perfect, but it’s improved in the last 20 years to produce better results and higher fidelity recordings. Today there are many audio formats, but they can be split into three groups based on fidelity and amount of compression.
Uncompressed Lossless Formats
Lossless audio files that have not been compressed in any way are the best audio format. None of the audio information has been lost to a compression algorithm. The most common lossless PCM (pulse-code modulation) audio formats are: WAV, AIFF, and AU.
Raw Headerless PCM
Raw headerless PCM is a digital audio that still exists in its raw pulse-code modulation form. This format has no metadata header information meaning there is no sample rate, bit depth, number of changes, or other information stored with the raw audio. If you want to listen to this format, you need to provide the header information for playback. This format is as unaltered as it gets and it’s how sound is captured for DVDs and CDs.
WAV
The WAV (Waveform Audio File Format) format is a lossless audio format. WAV files are essentially an audio wrapper, or container, for audio. It’s very popular today and is best when you need to retain the highest audio quality. Uncompressed WAV files are encoded with pulse code modulation, which is just sampling along an analog recording at regular intervals (referenced above).
AIFF
AIFF is the Mac counterpart to WAV. AIFF, which stands for Audio Interchange File Format, was developed in the late 1980s by Apple. Like WAV, it’s a wrapper that can hold multiple kinds of information, but is usually for uncompressed PCM audio.
Compressed Lossless Formats
Digital files can be compressed using a lossless algorithm to create a lossless format like FLAC, WMA, or ALAC. Lossless does not lose any information during compression because it shrinks only the file size, not the content of the audio file. This compression process works much like a ZIP or RAR file. Lossless formats retain the peaks and lows, but have much smaller file sizes for easy sharing and storage.
FLAC
The FLAC (Free Lossless Audio Codec) format is extremely popular because it can compress the original file by up to 50% without compromising the integrity of the audio. It’s also open-source and supports a range of sampling rates and bit depths that keep audio enthusiasts happy.
ALAC
ALAC is the lossless audio compression format created by Apple, hence the name Apple Lossless Audio Codec. All audio in the Apple Music catalog is encoded using ALAC. It’s comparable to FLAC although FLAC is slightly easier to unpack and playback. ALAC supports bit depths up to 32-bit and has a maximum sample rate of 384 kHz (384,000 samples per second!).
Compressed Lossy Formats
Lossy compression, on the other hand, removes information from the audio file itself. The peaks and lows are removed or clipped. Anything that is deemed unnecessary because it is at the fringes of human hearing. Lossy formats include MP3, OGG, and AAC.
MP3
MP3 stands for MPEG-1 Audio Layer-3. It is the most common compression format when the audio file needs to be very small. It was popular in the early 2000s but a trend towards higher-fidelity records has pushed it out of favor. MP3s are 1/12th the size of the original format but important audio information is lost during compression. The MP3 format supports sample rates from 16 to 48 kHz.
OGG
The OGG format, also known as Ogg Vorbis Compressed Audio File, is an open-source audio container format. The audio is saved in the Ogg container and compressed with the Vorbis audio compression algorithm. It’s very comparable in quality to MP3s.
AAC
The Advanced Audio Coding (AAC) format is essentially the successor to the MP3. It stores audio information in a similar sized file, with the .m4p or .m4a file extension, while retaining a higher sound quality. It can support up to 96 kHz sample rates and up to 48 channels.
WMA
WMA stands for Windows Media Audio. WMA is an audio file wrapper and audio codec. It’s used by the Windows Media Player that is built into all Windows operating systems. WMA is a lossy format, but there is a lossless version called Lossless WMA.
Lossy compression reigned supreme in the early days of digital music when the storage space available on hard drives was much less than it is today. While MP3s are better from a storage perspective, they are not ideal for listening. The lossy compression method introduces artefacts that create a sizzling sound. MP3s are also woefully flat sounding. Lossy formats don’t produce the best listening experiences and can leave listeners feeling worse than before listening. Lossless compression should be used wherever possible.
The Audio Reproduction Chain - The Sound Machine
Consumers listen to all modern music after it’s passed through the audio reproduction chain. New music is recorded using analog instruments and amplifiers that is immediately turned into digital information. Existing music is remixed or reproduced using a master record, either a master vinyl for older records or a master 32-bit 192 kHz digital recording.
From there, digital recordings are usually interpolated and downsampled and possibly compressed into more consumer-friendly formats. This is also when mixing and other alterations to the raw audio file are generally added. Most listeners use low-grade playback devices like headphones, earbuds, or simple stereo speakers that won’t reveal the benefits of a 32-bit 192 kHz recording. Instead, the industry standard is 44.1 kHz or 48 kHz, which is found on CD recordings and MP3s. Millions of albums have been released in the 44.1 kHz format.
When a digital file is finally in the hands of a consumer through their music library or a streaming service, it is converted back to an analog waveform using a DAC. This transforms the digital information, stored as 1s and 0s, into a voltage or current modulated electrical signal. The waveform is then transmitted through the amplifiers of headphones or speakers for listening.
There are various audio formats acting as different storage containers for digital audio information. Different formats have different ‘strengths’ and ‘weaknesses’ in that they are better or worse at storing information or easier or harder to unpack.
It’s difficult for the consumer to differentiate these differences after passing through the audio reproduction chain. But as a general rule, most of the issues or unwanted artefacts of the reproduction chain are added when downsampling, compressing or mixing digital audio. Ideally, audio files should be left as raw as possible to preserve the original characteristics. This means lossless audio formats are preferred and compressed lossy formats should be avoided for the best listening experiences.
Introduction
From the introduction of recorded audio in the early 1900s until the 1980s, everything was analog. The waveform was captured as a groove on vinyl or a magnetic signal on magnetic tape. But in 1982, that all changed with the introduction of the CD. Technological advances in digital music continued and today almost all consumer music today is recorded digitally. Digital music is available in various formats. Different formats offer different levels of quality, file size, and ease of encoding and decoding. When choosing the best format for listening, it’s a good idea to understand the basics of digital recording and the differences between the common audio file formats.
What is Sample Rate? What is Bit Depth? - Digital becomes Analog
With analog records, there is a direct waveform shape represented by a vinyl record's groove or the magnetization amount of a cassette tape. Analog is fantastic, but it does struggle with the issue of noise artifacts. When analog recording media is copied, more noise is introduced, and the overall fidelity is decreased. That’s why most records are created from a single master record.
But digital audio recording uses something called digital sampling. This reduces a continuous waveform signal into a discrete-time signal. A waveform is sampled at discrete and evenly-spaced intervals along the waveform, known as the sample rate. For example, a sample rate of 44.1 kHz captures 44,100 samples per second. 44.1 kHz is the industry standard for consumer records, but 48 kHz is also popular. Higher sample rates are used for recording (or audiophile projects), but later downsampled to make consumer CDs and other digital formats.
Each sample is given a precise number that represents the value of the signal at that point. The accuracy of this sample value is called the bit depth. For example, a 16-bit bit depth represents the sample value with a 16 digit binary (1s and 0s) number. 16-bit recording offers 65,535 audio amplitudes while 24-bit recording offers over 16 million values. The 32-bit float method is the latest and greatest way to store sample values. This method can be a bit confusing, but there are a few important things to note:
Float-point format – sample values stored with scientific notation, meaning they can be represented as very small and very large numbers. Much better than the fixed-point method.
Dynamic range – the possible dynamic range is increased to 1528 dB, up from 96 dB. The loudest sound possible is roughly 194-210 dB. This means the 32-bit method can support any and all sounds.
Large file sizes – using the 32-bit float for audio files means the files will be roughly 50% larger than before (with a 48 kHz sampling rate).
Avoid clipping or distortion – significantly more headroom means there’s no need to worry about the gain to avoid clipping and other distortions.
Digital records have the obvious benefit of being stored on digital storage formats like CDs or digital audio tapes (DAT). Digital records can also be used to listen on various electronic devices, including mp3 players, smartphones, and computers. But digital recordings have another benefit. They can be copied and mixed without introducing additional noise to the recording.
What is an Audio Format?
An audio format is a specific way to store audio. Different formats have different ‘strengths’ and ‘weaknesses’ in that they are better or worse at storing information or easier or harder to unpack. An audio format can be defined by its quality and fidelity.
The audio reproduction chain isn’t perfect, but it’s improved in the last 20 years to produce better results and higher fidelity recordings. Today there are many audio formats, but they can be split into three groups based on fidelity and amount of compression.
Uncompressed Lossless Formats
Lossless audio files that have not been compressed in any way are the best audio format. None of the audio information has been lost to a compression algorithm. The most common lossless PCM (pulse-code modulation) audio formats are: WAV, AIFF, and AU.
Raw Headerless PCM
Raw headerless PCM is a digital audio that still exists in its raw pulse-code modulation form. This format has no metadata header information meaning there is no sample rate, bit depth, number of changes, or other information stored with the raw audio. If you want to listen to this format, you need to provide the header information for playback. This format is as unaltered as it gets and it’s how sound is captured for DVDs and CDs.
WAV
The WAV (Waveform Audio File Format) format is a lossless audio format. WAV files are essentially an audio wrapper, or container, for audio. It’s very popular today and is best when you need to retain the highest audio quality. Uncompressed WAV files are encoded with pulse code modulation, which is just sampling along an analog recording at regular intervals (referenced above).
AIFF
AIFF is the Mac counterpart to WAV. AIFF, which stands for Audio Interchange File Format, was developed in the late 1980s by Apple. Like WAV, it’s a wrapper that can hold multiple kinds of information, but is usually for uncompressed PCM audio.
Compressed Lossless Formats
Digital files can be compressed using a lossless algorithm to create a lossless format like FLAC, WMA, or ALAC. Lossless does not lose any information during compression because it shrinks only the file size, not the content of the audio file. This compression process works much like a ZIP or RAR file. Lossless formats retain the peaks and lows, but have much smaller file sizes for easy sharing and storage.
FLAC
The FLAC (Free Lossless Audio Codec) format is extremely popular because it can compress the original file by up to 50% without compromising the integrity of the audio. It’s also open-source and supports a range of sampling rates and bit depths that keep audio enthusiasts happy.
ALAC
ALAC is the lossless audio compression format created by Apple, hence the name Apple Lossless Audio Codec. All audio in the Apple Music catalog is encoded using ALAC. It’s comparable to FLAC although FLAC is slightly easier to unpack and playback. ALAC supports bit depths up to 32-bit and has a maximum sample rate of 384 kHz (384,000 samples per second!).
Compressed Lossy Formats
Lossy compression, on the other hand, removes information from the audio file itself. The peaks and lows are removed or clipped. Anything that is deemed unnecessary because it is at the fringes of human hearing. Lossy formats include MP3, OGG, and AAC.
MP3
MP3 stands for MPEG-1 Audio Layer-3. It is the most common compression format when the audio file needs to be very small. It was popular in the early 2000s but a trend towards higher-fidelity records has pushed it out of favor. MP3s are 1/12th the size of the original format but important audio information is lost during compression. The MP3 format supports sample rates from 16 to 48 kHz.
OGG
The OGG format, also known as Ogg Vorbis Compressed Audio File, is an open-source audio container format. The audio is saved in the Ogg container and compressed with the Vorbis audio compression algorithm. It’s very comparable in quality to MP3s.
AAC
The Advanced Audio Coding (AAC) format is essentially the successor to the MP3. It stores audio information in a similar sized file, with the .m4p or .m4a file extension, while retaining a higher sound quality. It can support up to 96 kHz sample rates and up to 48 channels.
WMA
WMA stands for Windows Media Audio. WMA is an audio file wrapper and audio codec. It’s used by the Windows Media Player that is built into all Windows operating systems. WMA is a lossy format, but there is a lossless version called Lossless WMA.
Lossy compression reigned supreme in the early days of digital music when the storage space available on hard drives was much less than it is today. While MP3s are better from a storage perspective, they are not ideal for listening. The lossy compression method introduces artefacts that create a sizzling sound. MP3s are also woefully flat sounding. Lossy formats don’t produce the best listening experiences and can leave listeners feeling worse than before listening. Lossless compression should be used wherever possible.
The Audio Reproduction Chain - The Sound Machine
Consumers listen to all modern music after it’s passed through the audio reproduction chain. New music is recorded using analog instruments and amplifiers that is immediately turned into digital information. Existing music is remixed or reproduced using a master record, either a master vinyl for older records or a master 32-bit 192 kHz digital recording.
From there, digital recordings are usually interpolated and downsampled and possibly compressed into more consumer-friendly formats. This is also when mixing and other alterations to the raw audio file are generally added. Most listeners use low-grade playback devices like headphones, earbuds, or simple stereo speakers that won’t reveal the benefits of a 32-bit 192 kHz recording. Instead, the industry standard is 44.1 kHz or 48 kHz, which is found on CD recordings and MP3s. Millions of albums have been released in the 44.1 kHz format.
When a digital file is finally in the hands of a consumer through their music library or a streaming service, it is converted back to an analog waveform using a DAC. This transforms the digital information, stored as 1s and 0s, into a voltage or current modulated electrical signal. The waveform is then transmitted through the amplifiers of headphones or speakers for listening.
There are various audio formats acting as different storage containers for digital audio information. Different formats have different ‘strengths’ and ‘weaknesses’ in that they are better or worse at storing information or easier or harder to unpack.
It’s difficult for the consumer to differentiate these differences after passing through the audio reproduction chain. But as a general rule, most of the issues or unwanted artefacts of the reproduction chain are added when downsampling, compressing or mixing digital audio. Ideally, audio files should be left as raw as possible to preserve the original characteristics. This means lossless audio formats are preferred and compressed lossy formats should be avoided for the best listening experiences.
Introduction
From the introduction of recorded audio in the early 1900s until the 1980s, everything was analog. The waveform was captured as a groove on vinyl or a magnetic signal on magnetic tape. But in 1982, that all changed with the introduction of the CD. Technological advances in digital music continued and today almost all consumer music today is recorded digitally. Digital music is available in various formats. Different formats offer different levels of quality, file size, and ease of encoding and decoding. When choosing the best format for listening, it’s a good idea to understand the basics of digital recording and the differences between the common audio file formats.
What is Sample Rate? What is Bit Depth? - Digital becomes Analog
With analog records, there is a direct waveform shape represented by a vinyl record's groove or the magnetization amount of a cassette tape. Analog is fantastic, but it does struggle with the issue of noise artifacts. When analog recording media is copied, more noise is introduced, and the overall fidelity is decreased. That’s why most records are created from a single master record.
But digital audio recording uses something called digital sampling. This reduces a continuous waveform signal into a discrete-time signal. A waveform is sampled at discrete and evenly-spaced intervals along the waveform, known as the sample rate. For example, a sample rate of 44.1 kHz captures 44,100 samples per second. 44.1 kHz is the industry standard for consumer records, but 48 kHz is also popular. Higher sample rates are used for recording (or audiophile projects), but later downsampled to make consumer CDs and other digital formats.
Each sample is given a precise number that represents the value of the signal at that point. The accuracy of this sample value is called the bit depth. For example, a 16-bit bit depth represents the sample value with a 16 digit binary (1s and 0s) number. 16-bit recording offers 65,535 audio amplitudes while 24-bit recording offers over 16 million values. The 32-bit float method is the latest and greatest way to store sample values. This method can be a bit confusing, but there are a few important things to note:
Float-point format – sample values stored with scientific notation, meaning they can be represented as very small and very large numbers. Much better than the fixed-point method.
Dynamic range – the possible dynamic range is increased to 1528 dB, up from 96 dB. The loudest sound possible is roughly 194-210 dB. This means the 32-bit method can support any and all sounds.
Large file sizes – using the 32-bit float for audio files means the files will be roughly 50% larger than before (with a 48 kHz sampling rate).
Avoid clipping or distortion – significantly more headroom means there’s no need to worry about the gain to avoid clipping and other distortions.
Digital records have the obvious benefit of being stored on digital storage formats like CDs or digital audio tapes (DAT). Digital records can also be used to listen on various electronic devices, including mp3 players, smartphones, and computers. But digital recordings have another benefit. They can be copied and mixed without introducing additional noise to the recording.
What is an Audio Format?
An audio format is a specific way to store audio. Different formats have different ‘strengths’ and ‘weaknesses’ in that they are better or worse at storing information or easier or harder to unpack. An audio format can be defined by its quality and fidelity.
The audio reproduction chain isn’t perfect, but it’s improved in the last 20 years to produce better results and higher fidelity recordings. Today there are many audio formats, but they can be split into three groups based on fidelity and amount of compression.
Uncompressed Lossless Formats
Lossless audio files that have not been compressed in any way are the best audio format. None of the audio information has been lost to a compression algorithm. The most common lossless PCM (pulse-code modulation) audio formats are: WAV, AIFF, and AU.
Raw Headerless PCM
Raw headerless PCM is a digital audio that still exists in its raw pulse-code modulation form. This format has no metadata header information meaning there is no sample rate, bit depth, number of changes, or other information stored with the raw audio. If you want to listen to this format, you need to provide the header information for playback. This format is as unaltered as it gets and it’s how sound is captured for DVDs and CDs.
WAV
The WAV (Waveform Audio File Format) format is a lossless audio format. WAV files are essentially an audio wrapper, or container, for audio. It’s very popular today and is best when you need to retain the highest audio quality. Uncompressed WAV files are encoded with pulse code modulation, which is just sampling along an analog recording at regular intervals (referenced above).
AIFF
AIFF is the Mac counterpart to WAV. AIFF, which stands for Audio Interchange File Format, was developed in the late 1980s by Apple. Like WAV, it’s a wrapper that can hold multiple kinds of information, but is usually for uncompressed PCM audio.
Compressed Lossless Formats
Digital files can be compressed using a lossless algorithm to create a lossless format like FLAC, WMA, or ALAC. Lossless does not lose any information during compression because it shrinks only the file size, not the content of the audio file. This compression process works much like a ZIP or RAR file. Lossless formats retain the peaks and lows, but have much smaller file sizes for easy sharing and storage.
FLAC
The FLAC (Free Lossless Audio Codec) format is extremely popular because it can compress the original file by up to 50% without compromising the integrity of the audio. It’s also open-source and supports a range of sampling rates and bit depths that keep audio enthusiasts happy.
ALAC
ALAC is the lossless audio compression format created by Apple, hence the name Apple Lossless Audio Codec. All audio in the Apple Music catalog is encoded using ALAC. It’s comparable to FLAC although FLAC is slightly easier to unpack and playback. ALAC supports bit depths up to 32-bit and has a maximum sample rate of 384 kHz (384,000 samples per second!).
Compressed Lossy Formats
Lossy compression, on the other hand, removes information from the audio file itself. The peaks and lows are removed or clipped. Anything that is deemed unnecessary because it is at the fringes of human hearing. Lossy formats include MP3, OGG, and AAC.
MP3
MP3 stands for MPEG-1 Audio Layer-3. It is the most common compression format when the audio file needs to be very small. It was popular in the early 2000s but a trend towards higher-fidelity records has pushed it out of favor. MP3s are 1/12th the size of the original format but important audio information is lost during compression. The MP3 format supports sample rates from 16 to 48 kHz.
OGG
The OGG format, also known as Ogg Vorbis Compressed Audio File, is an open-source audio container format. The audio is saved in the Ogg container and compressed with the Vorbis audio compression algorithm. It’s very comparable in quality to MP3s.
AAC
The Advanced Audio Coding (AAC) format is essentially the successor to the MP3. It stores audio information in a similar sized file, with the .m4p or .m4a file extension, while retaining a higher sound quality. It can support up to 96 kHz sample rates and up to 48 channels.
WMA
WMA stands for Windows Media Audio. WMA is an audio file wrapper and audio codec. It’s used by the Windows Media Player that is built into all Windows operating systems. WMA is a lossy format, but there is a lossless version called Lossless WMA.
Lossy compression reigned supreme in the early days of digital music when the storage space available on hard drives was much less than it is today. While MP3s are better from a storage perspective, they are not ideal for listening. The lossy compression method introduces artefacts that create a sizzling sound. MP3s are also woefully flat sounding. Lossy formats don’t produce the best listening experiences and can leave listeners feeling worse than before listening. Lossless compression should be used wherever possible.
The Audio Reproduction Chain - The Sound Machine
Consumers listen to all modern music after it’s passed through the audio reproduction chain. New music is recorded using analog instruments and amplifiers that is immediately turned into digital information. Existing music is remixed or reproduced using a master record, either a master vinyl for older records or a master 32-bit 192 kHz digital recording.
From there, digital recordings are usually interpolated and downsampled and possibly compressed into more consumer-friendly formats. This is also when mixing and other alterations to the raw audio file are generally added. Most listeners use low-grade playback devices like headphones, earbuds, or simple stereo speakers that won’t reveal the benefits of a 32-bit 192 kHz recording. Instead, the industry standard is 44.1 kHz or 48 kHz, which is found on CD recordings and MP3s. Millions of albums have been released in the 44.1 kHz format.
When a digital file is finally in the hands of a consumer through their music library or a streaming service, it is converted back to an analog waveform using a DAC. This transforms the digital information, stored as 1s and 0s, into a voltage or current modulated electrical signal. The waveform is then transmitted through the amplifiers of headphones or speakers for listening.
There are various audio formats acting as different storage containers for digital audio information. Different formats have different ‘strengths’ and ‘weaknesses’ in that they are better or worse at storing information or easier or harder to unpack.
It’s difficult for the consumer to differentiate these differences after passing through the audio reproduction chain. But as a general rule, most of the issues or unwanted artefacts of the reproduction chain are added when downsampling, compressing or mixing digital audio. Ideally, audio files should be left as raw as possible to preserve the original characteristics. This means lossless audio formats are preferred and compressed lossy formats should be avoided for the best listening experiences.
There are various audio formats acting as different storage containers for digital audio information. Ideally, audio files should be left as raw as possible to preserve the original characteristics.
There are various audio formats acting as different storage containers for digital audio information. Ideally, audio files should be left as raw as possible to preserve the original characteristics.
There are various audio formats acting as different storage containers for digital audio information. Ideally, audio files should be left as raw as possible to preserve the original characteristics.
There are various audio formats acting as different storage containers for digital audio information. Ideally, audio files should be left as raw as possible to preserve the original characteristics.
VISUAL ACOUSTIC EXPERIENCE
VISUAL ACOUSTIC EXPERIENCE
Cutting-edge startup redefining sensory experiences. We create unparalleled technology for immersion in auditory landscapes.
Meet our blog author, a blockchain enthusiast and fintech expert with a passion for sharing insights on decentralized finance trends.
Music from the different perspective
Don't miss a beat - experience musical bits from the articles below.
View More
Psychology
Understanding the altered perception of music while on LSD sheds light on the broader relationship between psychedelics and sensory perception. It raises questions about the mind's ability to perceive reality, the flexibility of our sensory processing, and how deeply music is woven into the human experience.
Read
Psychology
Understanding the altered perception of music while on LSD sheds light on the broader relationship between psychedelics and sensory perception. It raises questions about the mind's ability to perceive reality, the flexibility of our sensory processing, and how deeply music is woven into the human experience.
Read
Psychology
Understanding the altered perception of music while on LSD sheds light on the broader relationship between psychedelics and sensory perception. It raises questions about the mind's ability to perceive reality, the flexibility of our sensory processing, and how deeply music is woven into the human experience.
Read
Metaphysics
Brain music is an intriguing intersection of neuroscience and auditory experience. It refers to a variety of phenomena where the human brain interacts with music, whether it be the neurological impacts of listening to music or the sonification of brain waves into audible frequencies.
Read
Metaphysics
Brain music is an intriguing intersection of neuroscience and auditory experience. It refers to a variety of phenomena where the human brain interacts with music, whether it be the neurological impacts of listening to music or the sonification of brain waves into audible frequencies.
Read
Metaphysics
Brain music is an intriguing intersection of neuroscience and auditory experience. It refers to a variety of phenomena where the human brain interacts with music, whether it be the neurological impacts of listening to music or the sonification of brain waves into audible frequencies.
Read
Metaphysics
When discussing sound healing, we often refer to specific frequencies that are believed to have particular benefits. For instance, the Solfeggio frequencies, a series of six tones that date back to early sacred music, are claimed to have properties ranging from repairing DNA to opening the heart chakra.
Read
Metaphysics
When discussing sound healing, we often refer to specific frequencies that are believed to have particular benefits. For instance, the Solfeggio frequencies, a series of six tones that date back to early sacred music, are claimed to have properties ranging from repairing DNA to opening the heart chakra.
Read
Metaphysics
When discussing sound healing, we often refer to specific frequencies that are believed to have particular benefits. For instance, the Solfeggio frequencies, a series of six tones that date back to early sacred music, are claimed to have properties ranging from repairing DNA to opening the heart chakra.
Read
Metaphysics
Solfeggio frequencies are ancient tones believed to have healing properties. This article delves into their history, from Gregorian Chants to modern rediscovery, examines their effects on emotional and physical well-being, and scrutinizes the scientific research behind these mysterious frequencies.
Read
Metaphysics
Solfeggio frequencies are ancient tones believed to have healing properties. This article delves into their history, from Gregorian Chants to modern rediscovery, examines their effects on emotional and physical well-being, and scrutinizes the scientific research behind these mysterious frequencies.
Read
Metaphysics
Solfeggio frequencies are ancient tones believed to have healing properties. This article delves into their history, from Gregorian Chants to modern rediscovery, examines their effects on emotional and physical well-being, and scrutinizes the scientific research behind these mysterious frequencies.
Read
Cutting-edge startup redefining sensory experiences. We create unparalleled technology for immersion in auditory landscapes.
Copyright ©2024 VA Visual Acoustic Technologies GmbH. All rights reserved.
Cutting-edge startup redefining sensory experiences. We create unparalleled technology for immersion in auditory landscapes.
Copyright ©2024 VA Visual Acoustic Technologies GmbH. All rights reserved.
Cutting-edge startup redefining sensory experiences. We create unparalleled technology for immersion in auditory landscapes.
Copyright ©2024 VA Visual Acoustic Technologies GmbH. All rights reserved.
Cutting-edge startup redefining sensory experiences. We create unparalleled technology for immersion in auditory landscapes.
Copyright ©2024 VA Visual Acoustic Technologies GmbH. All rights reserved.