The records that define 1970s Ghanaian highlife were made with no compressor, no reverb unit and no outboard of any kind. A peer-reviewed study of the Ghana Film Studio puts it flatly: "the available signal processor at the time was solely the mixing console's EQ knobs and volume faders." That is the entire processing rack. Two EQ knobs per channel and a row of faders.
So if you want to know how highlife was recorded, the answer is awkward for a shop that sells outboard for a living. It was recorded by a man named Francis Kwakye, who knew where to put a microphone, knew what to take out at 300Hz, and had nothing else to fall back on. This is the best argument against buying more gear that we are ever likely to publish, and it is made entirely out of gear detail.
Francis Kwakye, engineer, trumpeter, head of sound
Kwakye was born in Kumasi in 1948 and trained as a trumpeter in Nkrumah's Young Pioneers. That matters more than it sounds. He came to the desk as a horn player, from a movement that put ensembles on stages, not as a technician who learned music afterwards. He joined the Ghana Film Industry Corporation in 1969 and was Head of Sound by the mid-1970s.
The discography is the argument. Dr K. Gyasi and the Noble Kings, Sikyi Highlife (1974). 4th Dimension (1973). Uhuru Dance Band. C.K. Mann. Ebo Taylor (1977). Alhaji K. Frimpong (1977). Alex Konadu (1976). If you own a highlife reissue with that dry, forward, live-in-the-room sound, there is a good chance Kwakye cut it, with less signal processing than sits in most bedrooms in Ballito.
The Ghana Film Studio signal chain, decade by decade
The chain changed three times in about fifteen years, and each change is worth naming precisely, because the precision is the only thing protecting this story from becoming folklore.
| Period | Capture | Desk | Notes |
|---|---|---|---|
| 1960s | Nagra portable, single track | None documented | Location and field work as much as studio work |
| Late 1960s | Telefunken Magnetophon, 2-track mono | Neve (source says "12-track", read 12-channel) | Balance decided on the way in, not afterwards |
| 1970s | Same tape machine, expanded microphone locker | Same Neve | AKG condensers on percussion and drum overheads; Neumann condensers on winds and backing vocals; Beyer dynamics on bass and lead guitar |
| Late 1970s | Source says "8-track Revox" | Not documented | Revox never built an 8-track machine. See below. |
Note what is missing from that table: model numbers. The microphone brands are documented, the models are not. Nobody wrote down whether the AKG on the congas was a C12, a C28 or a D19, and we are not going to tell you it was. A gear shop that invents a model number to make a story tidier has nothing left to sell you. The brands are real, the models are gone.
A 2-track mono Telefunken Magnetophon is the important line in that table. Two tracks, mono. You are not recording a band to twelve tracks and sorting it out later. You are placing eight or ten musicians in a room, balancing them on the desk in real time, and printing the result. Every decision is permanent at the moment it is made. That is a different craft from the one most people mean when they say mixing.
Two things in the sources that cannot be right
The "12-track" Neve
The source describes a "12-track Neve desk". Neve did not build a 12-track desk, because desks are not measured in tracks; tape machines are. What Neve did build, in quantity, in that era, were 12-channel and 16-channel consoles. The overwhelmingly likely reading is a 12-channel Neve feeding a 2-track machine, which fits the rest of the account exactly: twelve inputs for a full dance band, two tracks out. We think the source means 12-channel. We cannot prove it, so we are flagging it rather than quietly correcting it.
The "8-track Revox"
The late-1970s upgrade is given as an "8-track Revox". Revox never made an 8-track. Revox built superb 2-track and quarter-track machines, the A77, the B77, the PR99, and the professional side of the business under the Studer name built the multitracks. An 8-track in a Ghanaian studio at the end of the 1970s is entirely plausible; a Revox-badged one is not. It may have been a Studer that got remembered as a Revox, which happens constantly and is an easy slip given the two names sat in the same building in Regensdorf. It may have been another make entirely. We do not know, and neither does anyone who has written about it.
Both errors get copied forward every time someone summarises the story. That tells you something about the state of African studio documentation generally: the people who ran these rooms were interviewed decades later, often once, and the transcript became the historical record.
How highlife was recorded once the tape was rolling
Here is the payoff, and it is the part almost nobody repeats because it requires reading past the anecdotes. Kwakye did two specific, describable things.
First, he changed the vocal-to-instrument ratio from 1:2 to 1:1. In the older Ghanaian recordings the voice sat at roughly half the level of the band. Kwakye brought it level with the band. That single decision is most of what people are hearing when they say a 1970s highlife record sounds modern: the vocal is present, in front, competing with the horns rather than sitting behind them.
Second, he mixed in three deliberate frequency bands.
| Band | Range |
|---|---|
| Low | 80 to 300Hz |
| Mid | 400Hz to 4kHz |
| High | 5 to 14kHz |
Look at the gaps. The bands as recorded do not meet: nothing is claimed for 300 to 400Hz, or for 4 to 5kHz. Whether those are guard bands he deliberately kept clear, or simply the rounding of a man describing decades-old practice in an interview, the source does not say. We are not going to build a theory on it. But if you have ever pulled 300Hz out of a horn section to stop it clouding a vocal, you already understand why a man with two EQ knobs per channel would think in exactly those terms.
That is the whole method. Three bands, a 1:1 vocal balance, careful placement, a fader ride. No compression to hold the vocal in place, so the singer held it. No reverb to create depth, so the depth had to come from the room and from the distances between the microphones and the players.
The room, and why it had a booth big enough for an audience
The Ghana Film Studio room was designed by a Canadian, Bill O'Neill. Its most interesting feature was a booth large enough to hold a studio audience, because Kwakye deliberately recreated concert conditions. Highlife bands played to dancers. Recorded in a silent room, to nobody, they played differently and worse. So he put people in the building.
That is an acoustic decision disguised as a social one, and no amount of outboard substitutes for it. The room later became TV3's Studio B.
Essiebons records were not recorded at Essiebons
This one needs correcting every time it comes up. Essiebons was Dick Essilfie-Bondzie's label and pressing operation, founded in 1969 in partnership with Philips. It released Ghana's first indigenously produced 33rpm album in 1970 and went on to issue roughly 800 singles and 200 LPs. It was not a recording studio.
Essiebons pressed and distributed. The records were cut at Ghana Film. Most online write-ups collapse the two, and you will see the Essiebons name attached to engineering decisions it had nothing to do with. If the credits on a highlife record say Essiebons, that tells you who pressed and sold it, not who put the microphones up.
Bokoor Studios, John Collins and a Tascam 4-track
The second half of this story is smaller, later and in some ways more extreme. John Collins founded Bokoor in Accra in 1982 and ran it into the late 1990s. In a 1993 interview he described the setup:
I've got a 10 channel mixer, but I basically use a 4-track recorder... I've got a Tascam 4-track cassette unit.
A four-track cassette machine. On that, and the ten-channel mixer feeding it, Collins recorded somewhere between 200 and 300 bands. The sources disagree and we are not going to split the difference and present it as fact; call it two to three hundred. Plus nine records and around sixty commercial cassettes.
The cassette part is not an eccentricity, it is economics. Collins on why:
The record industry collapsed here, around about the early '80s, and it switched to cassettes.
Vinyl pressing in Ghana stopped being viable, the market moved to cassette, and a studio recording to cassette was recording to the format its customers actually bought. Four tracks and a stereo master is the correct tool for that market. It is also a very thin pipe to push a full band through, and the records exist anyway.
Collins also flagged what was being lost, in the same interview:
All the studios now, except mine, and Ghana Films, no longer do live recordings.
That is the through-line between the two halves of this article. Ghana Film and Bokoor were the two rooms still tracking bands live, decades apart, on wildly different budgets, for the same reason: the music is an ensemble music and it stops working when you take the ensemble apart.
Collins founded the BAPMAF archive in 1990. It holds around 1,500 hours of music and 780 highlife 78rpm shellacs. In 2019 he and Dr Colter Harper digitised 100 hours of Bokoor master tapes for the Nketia archive at the University of Ghana. Those tapes survived forty years in a coastal climate well enough to be worth digitising, which is more than can be said for most of the cassettes that come through here.
Ambassador Records, Kumasi
Ambassador Records was founded in Kumasi in 1954 by A.K. Badu and is usually billed as Ghana's first record company. No equipment is documented anywhere we could find: no desk, no machines, no microphones, no room description. That is the entire honest entry. Every account that gives you a gear list for Ambassador has made it up, and we are not going to join in.
What vintage African recording techniques are worth to you now
The temptation is to turn this into a moral about how nobody needs equipment. That is not the lesson, and it would be a strange lesson from us.
The lesson is narrower. Kwakye's results came from the parts of the chain that sit before the processing: the room, the placement, the players, the balance decided on the way in. Those are the parts most people skip, because they are the parts you cannot buy. The three-band thinking and the 1:1 vocal balance are free, and you can apply both this afternoon.
Equipment makes those decisions easier to execute; it does not make them for you. If you want the same discipline with the same shape of tools, what is worth owning is a desk with real EQ on every channel, a clean tape path, and a way to commit. The Chilton M10/2 Mk6 is the closest thing we stock to the class of discrete British console Kwakye would have been sitting behind, and the Studer 169 is the same idea from the Swiss broadcast side. For the one-machine location approach, the Sony TC-510-2 is the machine reporters carried when they could not afford a Nagra, and the Akai M-7 is a valve-era reel-to-reel from the same decade as the Magnetophon. If the Revox question above interests you, the Revox PR-99 is what Revox actually built for professional use, and it is two tracks, not eight.
For the Bokoor end of the story, the small-desk-and-four-tracks approach, the TEAC Tascam M-30 and the Tascam TEAC Model 3 are the direct descendants of what Collins had, the TEAC AX-300 gives you Tamura-transformer Japanese iron in front of them, and the TEAC AN-300 is the noise reduction unit that made four tracks of cassette survivable in the first place. If you want the valve version of two EQ knobs and a fader, the Sansui MX-10 is a 1960 channel strip from before Sansui discovered transistors. For the three-band approach done deliberately, the Speck ASC is a four-band parametric with Jensen iron and enough resolution to find Kwakye's 300Hz.
Everything above is sold serviced, which is the part that costs money on gear this old. If you are reading this from Accra, Lagos or anywhere else on the continent, our guide to importing vintage studio gear into Africa covers duty, VAT, voltage and the shipping routes that do not lose things.
This piece is one of four we have written on African studio history. The others cover BOP Studios in Mmabatho, where Fela Kuti actually recorded, and how the continent got multitrack. Ghana Film sits early in that timeline and Bokoor sits near the end of it.
What we could not verify
- Microphone models. AKG, Neumann and Beyer are documented as the brands at Ghana Film in the 1970s. No model numbers survive in any source we found. We have not guessed.
- The Neve. The source says "12-track". We believe it means a 12-channel console. We cannot prove it and have flagged it rather than silently correcting it.
- The "8-track Revox". Revox never built an 8-track. What the studio actually installed in the late 1970s is unknown to us.
- The Nagra model. "A portable Nagra, single track" is as specific as the record gets. Nagra built several such machines across that period.
- Bokoor's band count. Sources give 200 and 300. We have quoted the range rather than choosing.
- Bokoor's mixer. Collins says "10 channel mixer". No make or model is given in the interview.
- Ambassador Records. Founded 1954 by A.K. Badu in Kumasi. No equipment of any kind is documented. Anything you read to the contrary is invention.
- Exact dates. Kwakye joined Ghana Film in 1969 and was Head of Sound by the mid-1970s. The precise year of the promotion is not recorded.
Image credits
- Mid-1970s control room: Jamie Anstey, Wikimedia Commons, CC BY-SA 4.0. Not a Ghanaian room, and captioned as such.
- Ebo Taylor: Schorle, Wikimedia Commons, CC BY-SA 3.0.
- Kaneshie market, Accra: Fquasie, Wikimedia Commons, CC BY-SA 4.0.
- All equipment photographs are our own, of stock on our bench.