MP3
In the early 1980s, Karlheinz Brandenburg began working on digital music compression at Germany's University of Erlangen-Nuremberg. He focused his research on how people perceive music and what could be removed from audio without human ears noticing. By 1989, he had completed his doctoral work and moved to AT&T-Bell Labs as a postdoctoral researcher. There he collaborated with James D. Johnston, known as JJ, to refine compression algorithms. The MP3 format emerged from merging two distinct technologies: ASPEC and MUSICAM. ASPEC came from a group including Fraunhofer Gesellschaft, AT&T, CNET, and Thomson. It offered high coding efficiency but was initially rejected for being too complex to implement in hardware. MUSICAM originated from Matsushita, CCETT, ITT, and Philips. This codec used integer arithmetic and sub-band filter banks designed for Digital Audio Broadcasting. In 1991, representatives from both groups met to combine their strengths. They created a hybrid system that kept the robust frame structure of MUSICAM while integrating the advanced psychoacoustic model from ASPEC. Brandenburg later joined the Fraunhofer Institute for Integrated Circuits in Erlangen in 1993. His team included Bernhard Grill and four other researchers who became known as The Original Six. Their combined effort produced the first practical implementation of an audio perceptual coder using Motorola 56000 DSP chips. This hardware-based solution proved far more efficient than earlier attempts by M.A. Krasner, whose equipment was too slow for real-world use.
The core principle behind MP3 relies on auditory masking, a phenomenon where one sound makes another inaudible. American physicist Alfred M. Mayer reported this effect in 1894 when he found that a tone could be rendered silent by another tone of lower frequency. Richard Ehmer expanded this research in 1959 with complete sets of auditory curves describing the phenomenon. Between 1967 and 1974, Eberhard Zwicker studied tuning and masking within critical frequency bands at Bell Labs under Harvey Fletcher. These foundational studies informed modern compression techniques used decades later. Perceptual coding first appeared in speech compression through linear predictive coding developed by Fumitada Itakura and Shuzo Saito in 1966. Bishnu S. Atal and Manfred R. Schroeder proposed adaptive predictive coding in 1978, which exploited human ear masking properties to reduce data size significantly. Ernst Terhardt constructed an algorithm describing auditory masking with high accuracy in 1982. The discrete cosine transform proposed by Nasir Ahmed in 1972 evolved into the modified discrete cosine transform (MDCT) by J.P. Princen, A.W. Johnson, and A.B. Bradley in 1987. This MDCT became a core component of the MP3 algorithm. Brandenburg tested his compression model using Suzanne Vega's song Tom's Diner because its nearly monophonic nature made imperfections easy to hear during playback. He listened to the track repeatedly while refining the algorithm until it preserved her voice without unnatural artifacts. Instrumental music proved easier to compress than vocal tracks due to predictable patterns and fewer transient sounds.
The Moving Picture Experts Group designed MP3 as part of its MPEG-1 standard suite approved in 1991. Two proposals competed for inclusion: MUSICAM and ASPEC. MUSICAM won selection due to its simplicity and error robustness despite offering lower coding efficiency than ASPEC. The editing process involved Leon van de Kerkhof from the Netherlands, Gerhard Stoll from Germany, and Yves-François Dehery from France. They worked under Professor Hans Musmann who chaired the ISO MPEG Audio group for several years. Karlheinz Brandenburg joined this working group along with James D. Johnston from the United States. Together they integrated ideas from both competing systems into what would become Layer III. The algorithms were finalized in 1992 and published as international standard ISO/IEC 11172-3:1993 in 1993. An extension called MPEG-2 Audio followed in 1995 as ISO/IEC 13818-3:1995. This version added support for lower sample rates and multichannel audio up to 5.1 surround sound configurations. A reference simulation software written in C language was developed between 1991 and 1996 by committee members. It produced bit-compliant files known as ISO 11172-5. This document became a freely available technical report approved in November 1994 and printed as CD 11172-5 in April 1994. Final publication occurred in 1998 as ISO/IEC TR 11172-5:1998. The first practical encoder l3enc arrived on the 7th of July 1994, released by Fraunhofer Society. File extensions changed from .bit to .mp3 on the 14th of July 1995. WinPlay3 player launched the 9th of September 1995 enabling real-time playback on personal computers.
A hacker nicknamed SoloH discovered the source code of dist10 MPEG reference implementation on University of Erlangen servers shortly after its release. He integrated it within a graphical interface allowing easy conversion from CD or WAV files. This software started widespread CD ripping and digital music distribution over the internet. Nullsoft's audio player Winamp released in 1997 gained popularity rapidly with an estimated 80 million active users still present in 2023. Windows Media Player 5.2 added MP3 support in 1998 alongside portable solid-state digital audio players like MPMan developed by SaeHan Information Systems headquartered in Seoul South Korea. Rio PMP300 followed sales despite legal suppression efforts by Recording Industry Association of America. In November 1997 website mp3.com offered thousands of MP3s created by independent artists for free. Small file sizes enabled peer-to-peer sharing networks previously impossible due to bandwidth limitations. Napster launched in 1999 as the first large-scale peer-to-peer filesharing network. Major record companies argued that free sharing reduced sales and labeled it music piracy. They pursued lawsuits against Napster which eventually shut down before being sold later returning as legitimate streaming service. Unauthorized sharing continued on next-generation networks while authorized services like Amazon.com Beatport Bleep eMusic Juno Records and reincarnated Napster began selling unrestricted music in MP3 format.
The small size of MP3 files allowed multiple albums worth of music to fit onto home computers with hard drives around 500 to 1000 megabytes during the mid-1990s. This compression was essential compared to MIDI notation or tracker files combining notation with short recordings of single instrument notes. Portable media players emerged including smartphones supporting near-universal adoption of the format. The Rio PMP300 became one of the earliest commercial devices selling after MPMan released in 1998 despite legal challenges from RIAA. By 2004 Sony introduced native MP3 support to its Walkman players following criticism and lower than expected sales of their proprietary ATRAC codec used in MiniDisc format. Earlier systems lacked fast forwarding and rewinding controls found on modern boomboxes. Bit reservoir technology enabled variable bit rate encoding allowing temporary changes in effective bit rate even within constant streams. Internal handling increased encoding delay but improved overall quality significantly. Decoders became more computationally efficient as CPU clock rates transitioned from MHz to GHz over time. Most decoders remained bitstream compliant producing decompressed output matching mathematical specifications within rounding tolerance defined by ISO/IEC 11172-3 standard.
Technicolor formerly called Thomson Consumer Electronics claimed control over MP3 licensing in many countries including United States Japan Canada and EU nations. They actively enforced these patents generating approximately €100 million for Fraunhofer Society in 2005 alone. In September 1998 Fraunhofer Institute sent letters requiring developers to obtain licenses distributing or selling encoders and decoders claiming unlicensed products infringed patent rights. This situation prevented LAME MP3 encoder project from offering official binaries usable directly on computers though unofficial compiled versions existed elsewhere. Sisvel S.p.A. Luxembourg-based company administers licenses applying to MPEG Audio alongside its US subsidiary Audio MPEG Inc. They sued Thomson resolving disputes in November 2005 granting Thomson a license to their patents. Motorola signed with Sisvel December 2005 licensing MP3-related patents except three exceptions expiring February 2017 and the 9th of April 2017. German officials seized MP3 players from SanDisk booth at IFA show Berlin September 2006 after Italian patents firm won injunction on behalf of Sisvel against SanDisk dispute over licensing rights. Texas MP3 Technologies sued Apple Samsung Electronics and Sandisk eastern Texas federal court February 2007 claiming infringement portable player patent assigned to them. All three companies settled claims January 2009. Alcatel-Lucent asserted several coding compression patents allegedly inherited from AT&T-Bell Labs suing Microsoft November 2006 for alleged infringement seven patents. San Diego jury awarded Alcatel-Lucent $1.52 billion damages the 23rd of February 2007 before court revoked award finding one patent not infringed another co-owned by AT&T and Fraunhofer licensed to Microsoft.
Advanced Audio Coding AAC designed as successor to MP3 remains most widely used alternative format today. Other lossy formats include mp3PRO and MP2 members same technological family depending roughly similar psychoacoustic models and MDCT algorithms. AAC uses purely MDCT approach significantly improving compression efficiency compared hybrid coding method combining MDCT with FFT employed by MP3. Many basic patents underlying these formats held by Fraunhofer Society Alcatel-Lucent Thomson Consumer Electronics Bell Dolby LG Electronics NEC NTT Docomo Panasonic Sony Corporation ETRI JVC Kenwood Philips Microsoft and NTT. Microsoft created competing standard Windows Media Audio WMA claiming superiority over MP3 during early digital audio player market growth phase. Open compression formats like Opus and Vorbis OGG available free charge without known patent restrictions offer alternatives to proprietary solutions. Lossless formats such as FLAC Free Lossless Audio Codec Apple Lossless provide unaltered audio content though increased file size compared lossy compression methods. MPEG-2 Audio Layer III added support for lower sample rates half those originally defined in MPEG-1 reducing frequency fidelity while cutting bit rate 50 percent. MPEG-2.5 extended ideas further named unofficially since MPEG-3 already had different meaning developed at Fraunhofer IIS reducing frame sync field from 12 bits to 11 bits widening scope including human speech applications requiring only 25% bandwidth possible using MPEG-1 sampling rates. Sample rate comparison table shows nine variations total across three generations supporting 8 kHz through 48 kHz ranges depending application needs.
Common questions
Who invented the MP3 format and when was it developed?
Karlheinz Brandenburg developed the MP3 format while working at Germany's University of Erlangen-Nuremberg in the early 1980s. He completed his doctoral work by 1989 and collaborated with James D. Johnston to refine compression algorithms before joining the Fraunhofer Institute for Integrated Circuits in 1993.
What is the core principle behind how MP3 compresses audio files?
The core principle behind MP3 relies on auditory masking, a phenomenon where one sound makes another inaudible. This effect was first reported by American physicist Alfred M. Mayer in 1894 and later expanded by Richard Ehmer in 1959 to describe complete sets of auditory curves.
When did the MP3 standard get officially approved and published?
The Moving Picture Experts Group designed MP3 as part of its MPEG-1 standard suite which was approved in 1991. The algorithms were finalized in 1992 and published as international standard ISO/IEC 11172-3:1993 in 1993.
Which software tools enabled widespread MP3 distribution and playback in the late 1990s?
Nullsoft's audio player Winamp released in 1997 gained popularity rapidly with an estimated 80 million active users still present in 2023. Windows Media Player 5.2 added MP3 support in 1998 alongside portable solid-state digital audio players like MPMan developed by SaeHan Information Systems headquartered in Seoul South Korea.
How much money did Fraunhofer Society generate from MP3 licensing in 2005?
Technicolor formerly called Thomson Consumer Electronics claimed control over MP3 licensing in many countries including United States Japan Canada and EU nations. They actively enforced these patents generating approximately €100 million for Fraunhofer Society in 2005 alone.