BINAURAL
120 YEARS OF HEADPHONE KINGDOM

n my coverage of the 98. Cracow Sonic Society Meeting I mentioned Qsound – a technique improving spacial effects in stereo recordings intended for loudspeakers playback. Its effects might be really surprising, but what's equally important is that when one listens to such recordings using headphones all the spacial relations within the recording also seem OK. When using cans one won't really hear any benefits from Qsound but sound quality won't be worsen either. There is a special technique though, developed for headphone music playback – such recordings are called binaural, and they are not the best choice for loudspeaker playback. It is a story about them.

The beginning

The first headphones were not designed for music at all, they were used by phone operators in 1870ties and 1880ties. It was not until 1881 that Théâtrophone Clementa Adera system was presented in France, that, starting in 1890, was used for listening to musical and theatrical spectacles from Opera Garnier over phone lines.

A similar invention was commercialized in Great Britain. Thanks to the Electrophone system, in 1895 folks could start rocking out to the sick beats of the local opera house from the comfort of their own home. Subscribers to the pricey service would listen through headphones that looked more like stethoscopes than a modern offerings as very large people produced very big sound on a stage miles away. Apart from music subscriber could also listen to a mess from local church (more HERE). These facts mean that this year we should celebrate the 120th anniversary of headphone presence in music fans homes.

Headphones vs loudspeakers

Listening to music using headphones differs significantly from listening using loudspeakers (or horns in the past). Cans allow to eliminate any influence of the room one listens to music in, and also a specific changes in the signal caused by ITD (interaural time difference) and ILD (interaural level difference). These are phase shifts and changes of sound level caused by a distance between ears and anatomical structure of our nose and ear.

It is about so called head-related transfer function (HRTF). Short version – when we listen to music using loudspeakers each ear can hear both, right and left channel – the left ear receives directly sound from left channel but also from the right channel changed by HRTF, and the right ear receives directly sound from right channel and HRTF modified sound from left channel (more HERE).

So when we listen to music using loudspeakers the way were are build influences the sound we can hear. This problem is gone when we listen to music using headphones but it also affects spacial relations in music causing a specific sound presentation described by people not used to headphones as „sound inside the head” which some simply don't like. But there is a way out of it – binaural recordings (Latin for ‘twofold, two apiece’). An interesting fact – today we call a two-channel sound: stereo, although once stereo and binaural were synonyms. Today ‘stereophonic’ means 'two-channel' for both, loudspeaker and headphone listening and 'binaural’ is used exclusively for special recordings intended for headphone listening.

Binaural

The English Wikipedia dates history of binaural music playback back to 1881 to the first binaural unit called the Theatrophone, same information can be found on Bowers & Wilkins blog (more HERE). Historically speaking it is true as back then there was no real difference between 'stereo' and 'binaural' listening. Only later works by Alan Blumlein made the distinction clearer. Since then ‘Stereo’ meant exactly the same as it does today, meaning a two-channel, spatial presentation. So by today's understanding these first transmissions were stereophonic and not binaural – the distinctions comes from different setup of microphones for particular type of recording.

„Dummy head” by Neumann, model KU80. Image by Neumann.

The difference between mono, stereo, multi-channel and binaural recordings lays in different microphones arrangement during recording. Two high-fidelity microphones are used for binaural recordings that are placed inside so called 'dummy head'. It looks like actual head with ears and nose. Microphones are inset in ear-shaped molds to fully capture all of the audio frequency adjustments known as head-related transfer functions (HRTFs). A regular stereophonic recording creates space that is modified by a listener while listening. Stereo recordings might also use two microphones (usually they use more), but they are simply set apart and no 'dummy head' is used.

After a while when Clement Ader's invention lost its charm of a novelty people realized how uncomfortable was listening using headphones which was the reason why efforts were focused rather on developing new solutions for loudspeakers and not for cans. It wasn't until the 1950ties that a radio station in Connecticut started binaural broadcast. An interesting fact about it – at the time there was no stereophonic radio yet, so the left channel was broadcast using one frequency and the right one using another. To listen to both channels one needed two identical radios each of them receiving different station.

Popularization

At first using binaural recordings was rather costly. Also creating such recordings was expensive – the equipment necessary to realize such recordings was very pricey. Models of 'dummy head' that were used most often were Neumann's KU80 and KU100. One can find their characteristic shape on many photos and their name on many album covers, not all of them intended for headphone listening. For example Telarc used KU100 for many years to improve spatial effects in their recordings and to make possible listening to stereophonic recordings on Dolby Surround systems.

„Dummy head” by Neumann, model KU100, used by Telarc. Image by Neumann.

It was Manfred Schunke and Americans living in Berlin – Ed and Mary Kay who introduced this effect to popular music. The Kunstkopf-Stereophonie concept (Kunstkopf = dummy head) was applied by them on many krautrock albums. Schunke was responsible for the sound of Can group recordings (Flow Motion, 1972), Code III (Planet of Man, with support of Klaus Schulze), but also for Lou Reed's Street Hassle (1978). The latter is often pointed out as the first commercial use of binaural technique for popular music.

Code III, Planet of Man, Delta Acoustic/Wah Wah Records LPS135, 180 g LP (1974/2014)

Binaural effects were used also on many different albums including Alan Parsons Project's Tales of Mystery and Imagination and Pink Floyd's The Final Cut (to be exact the Holophonic Zuccarelli Labs ltd system was used here). Binaural elements were mixed with traditional ones recorded on a multi-track system. In most cases the idea was to enrich studio recordings with realistic ambiance. Manfred Schunke's recordings were an exception as he used 'dummy head' for his experiments with sound. The achieved spacial effects were really impressive whether recordings were listened via headphones or loudspeakers.

Also Tchad Blake, an American music producer, sound engineer and musician, winner of two Grammy Awards for his productions, who was hired for recording and producing Pearl Jam's album Binaural (2000) decided to use some binaural elements. Blake was already known for his experiments with spatial effects – list of his clients looked like a Who is Who of musical world. He used a Neumann's modified KU100 when recording Binaural.

The basic goal of using binaural technique is to recreate realistic acoustics and placement of each instrument in the space. That is why it is most commonly used for classical music. Some time ago STAX , manufacturer of electrostatic headphones, released their own sampler Stax Raumklang CD, and now one can find on the Internet an album by Duo Seraphim called A Renaissance Journey (Pentafonie, 24 bits/352,8 kHz).

One of the most interesting examples of using binaural techniques come from Zenph Studios. Their portfolio includes only 5 albums, four of them recorded using a mechanical piano. Computer controlling piano used drivers created by Zenph. It is a „Re-Performance” of and actual recording performed in new acoustical environment. Using this technique they recorded „re-performed” albums by Art Tatum (SACD/CD), Oscar Peterson (CD), Glen Gould (SACD/CD) and Rachmaninoff (CD).

The recordings were done at the same time with two sets of microphones – one for the recording to be played using loudspeakers and the other, binaural one, recorded using dummy head for headphones. On one album there are tracks intended for loudspeaker listening followed for those intended for headphones.

• Glenn Gould, Bach: The Goldberg Variations, Sony BMG Music/Sony Classical/Zenph Studios 9703350-2, “Zenph Re-Performance”, SACD/CD (2007)
• Oscar Peterson, Unmistakable, Sony Music/Zenph Studios 951702, “Zenph Re-Performance”, CD (2011)
• Rachmaninoff Plays Rachmaninoff, RCA Red Seal/Sony Music 8697-48971-2, “Zenph Re-Perfomance”, CD (2009)
• Art Tatum, Piano Starts Here, Columbia/Sony Classical 97 22218 2, “Zenph Re-Performance”, SACD/CD (2008)

Sound

iPod changed the audio world, with significant help from headphones. Today more people use headphones than altogether throughout history before. So I can't really understand why aren't there more binaural recordings available, especially with live performances. These could be released on CD and SACD and also as downloadable files. I think this could be a real hit on the market.

Although… The very first contact with binaural recording might be bit confusing. When listening to „regular” recording with headphones sound seems to be rendered in the middle of the head with some spacial effects at sides. Each instrument is clearly separated and has proper weight. With binaural recordings instruments has less weight and are not that palpable. It might even seem that there is „less” sound.

Information regarding recording and system provided in the insert of Code III album, Planet of Man

One should keep on listening to binaural recordings for some time and then go back to regular ones; with Zenph recordings it is simple. Such a comparison clearly shows shortcomings of a stereophonic presentation. Placement of all elements in the space is not very precise and usually all of them are placed on a single line between left and right ear. Binaural recordings on the other hand offer real depth and are more nuanced. What we initially perceive as a „lack” of information turns out to be much richer with information, that in regular stereo recordings are „hidden” behind not so precisely defined phantom images.

Binaural recordings deliver also much richer timbre of instruments, that is smoother, deeper, more creamy. Listen to some binaural recordings with additional spacial effects like Code III and you'll be amazed by a realistic presentation of these effects, you will involuntarily look around you looking for sources of sounds coming from different directions.

Problems

Problem no. 1 is a very small number of binaural recordings available. There are much more so called binaural beats, intended for brain stimulation (meditation), but these have nothing to do with music. So, as already said, there are very few binaural recordings for music fans. That is why I appreciate so much releases by Telarc, Dolby Surround ones, and the ones from Zenph. I forgot – I love krautrock too – there are many examples of really good music with a realistic 3D spacing.

There are also some technical issues. It is not common knowledge but many headphones are designed with a large, even 10 dB dip at frequency around 5 kHz. The intention is to compensate for changes introduced to frequency range by HRTF. When listening using headphones such a compensation is not needed. If we used cans with flat frequency response they could sound too bright. Also each man differs with detail of anatomic structure so in fact there are no universal cans that would be best for everyone. Best headphones are tuned in such a way to sound best for most people.

On the other hand cans with flat frequency response would be best to listen to binaural recordings, but one won't find such headphones on the market. Maybe that's an opportunity at least some audio manufacturers will spot and offer sound processors that will allow users to adjust sound delivered by headphones to particular needs of each pair of ears, and to listening to classic stereo recordings and binaural ones. The first such product is the P-1 – headphone amplifier and sound processor in one body made by Ancient Audio.

Ancient Audio P-1 – headphone amplifier and sound processor in one body

Do you remember the coverage by Mark Robinson from Japan? This headphone fan has been using his computer for years as a sound processor for his cans. Let me also remind you that few years ago Dolby promoted their sound processing solution called Dolby Headphone. I had a chance to listen to such headphones and I have to admit that they sounded really good. Unfortunately I haven't heard any news on development of this solution for time now, so it seems it wasn't that successful.

Headphones + loudspeakers

Now, let me tell you about yet another interesting solution that is intended to let listener enjoy binaural experience using loudspeakers.

Mid 2012 the American label Chesky Records offered download of audio files coded using system called Binaural+. Playback required no special decoders and it was possible using both, headphones and loudspeakers.

For that to work signal had to prepared in a special way. The process was based on works of Dr. Edgar Choueiri of Princeton University. He developed theory for HRTF filters at Ambisonics Institute. Basing on his works Choueiri developed BACCH (Band-Assembled Crosstalk Cancellation Hierarchy) filter, used for coding Binaural+ recordings (more HERE). Today a special decoder called Bacch-SP is available that allows to decode binaural recordings in a way that allows to present similar effect using loudspeakers as the ones produced by headphones (more HERE).

At first only single files coded with Binaural+ were available (in three different resolutions), but after a while also whole CD albums were also released. The Binaural+ series includes at the moment 22 titles. Music for these albums is recorded digitally in high resolution (PCM 24/192) using 'dummy head' (Chesky calls it „Binaural head”). This particular „head” is called Lars.

When describing sound of Binaural+ recordings one has to considered a certain approach of Chesky Records towards sound all of its recordings. The most characteristic features is relation between direct sound and its reflections – Chesky recordings sport much more of the latter than most other recordings. No artificial reverb is added, the effect comes from placing microphones further away from sound source. This creates sensation of a more distant sound than usually, it feels like we sit in the middle of a large concert hall.

One can hear it also when listening with headphones. Spacing seems very natural especially when compared with „normal” headphone presentation that is quite „flat” and presented very close to us. The air surrounds us and sound sources seem to be placed rather far away from us. When listening to such recordings using loudspeakers it sounds like any other Chesky recording. That's already a success as the „regular” binaural recordings sound differently. The basic decision everybody has to make is whether he likes this type of presentation or not – not everybody will. If so – there are some nice recordings one may truly enjoy.

Alexis Cole, A Kiss In The Dark, Chesky Records JD366, “Chesky Records Binaural Series”, CD (2014)

Summary

This incredible popularity of headphones has not caused increase of binaural recordings popularity. The main obstacles are technical issues that companies who would want to records and distribute such recordings have to face, but also a limited demand for such recordings. There is not demand for such recordings as only very few people had a chance to experience advantages of high quality binaural recording, plus the music that is available in this form might not necessarily meet wide public expectations. Some that have a chance to listen to binaural recordings might not even be interested in them as they sound so differently from regular stereo ones. It's a real pity as only such recordings allow listener to have a real concert hall „on his head”.

Wojciech Pacuła
Chief Editor