> This article is the longest piece I’ve published on Bjango’s site, and it took a couple of years of research. I purchased around 25 pieces of music gear. I emailed Imogen Heap, and to my surprise, someone from her team got back to me and confirmed the exact harmonizer used on Hide and Seek.
> It’s been a huge effort, and I’m confident it contains a lot of information that is not widely known. For those of you who are into Daft Punk, I hope it’s interesting.
(They also did the Soundtrack for Tron:Legacy, basically their fifth studio album)
Anyway, I keep remembering how panned 'Human After All' was, and how bad the reviews were because the album was too "mechanical" and was "missing the warmth of House", while this is EXACTLY how the genre evolved in the years to come and none of those music experts saw this.
Many journalists did a retrospective of it a few years later and admitted that they misjudged it.
It's not that Daft Punk drove the industry in this direction, the album wasn't well-received by most at that time. They showed the destination of a journey while people didn't even realize they are traveling...
In the end, it appears that 'Random Access Memories' is one of their least innovative and "lasting" albums. It's probably their most successful one, the most complex to conceptualize and produce, but IMO it has the least unique character of all their productions.
Looking at the whole picture, the product of "Random Access Memories" is less the music, but the duo celebrating the process of production itself...
The reason that I dislike Human After All is that, quite simply, it's not fun.
It's dour. It's depressing. And it's repetitive in a way that feels tiresome; You can't dance to most of it or even really tap your foot to it.
What happened was that Daft Punk challenged themselves to make an album in six weeks and ended up with a showcase for a few neat guitar pedals and two fun songs (Robot Rock and Technologic), one of which consists almost entirely of a barely-changed sample (https://www.youtube.com/watch?v=PFwGQAEYqHs).
Electronic music certainly evolved in ways that made it less warm, but that didn't make it less fun. Case in point, Skrillex's early music is weird and playful despite relying on 'cold' synth sounds. Human After All is just cold.
"in a way that feels tiresome" is the operative phrase here. I like the throwaway tracks on Homework that are twice as long and half as interesting more than I like most of Human After All. They're fun.
Not to be that guy, but techno is only repetitive if you are a) listening to boring/uninspired techno (the same kind of complaint being leveled here against HAA), or b) not paying attention.
Fine not to like it for those reasons. But maybe try listen to it next time from a more psych, garage, or post-rock perspective. They were influenced very early on by groups like Spacemen 3, and I see HAA as them doing something big and dirgey in this kind of school. I even recall the liner notes saying only "all guitars by Daft Punk".
The sound design is certainly interesting from the perspective of psych, garage, post-rock, etc., but the sound design doesn't make up for the songwriting. You need both. Or at least, I do.
I think it's a just a more experimental record. The lack of songwriting is kind of the thesis of the entire album, right? (technology, repetition, etc) It's aims are not to be Digital Love or their more conventional song-based stuff. If anything, it's almost a return to their origins like some of the more techno-orientated tracks on Homework.
That's what I meant with what it has in common with garage and psych. Not the textures or sounds, but the relentlessly experimental approach to convention. Not saying it is perfectly executes, but I think this what it is meant to be.
What you’re describing is an album that’s more fun to think about than it is to listen to!
It definitely has a theme, but unfortunately as executed it’s basically “what if we defied convention by using live instruments and making a shitty album?”
When you bring up psychedelic, experimental post-rock, I want to mention that one of my favorite bands is Battles. They’re a good example of what you can do when you defy convention!
Ha. Fair, but music that makes us think more than feel isn't necessarily always a terrible thing. John Cage, Steve Reich, even Battles is pretty intellectual in a way (it's quite literally math rock lol).
I guess we like different artists for different reasons, and you like Daft Punk in boogie mode. Which I definitely get, as it is their strong suit. But I just think this is album is a bit of an anomaly in their catalogue and might require having to unlearn them a bit to dig it.
Their live album Alive 1997 would have to be my favourite of theirs btw. But I don't hate Human After All as many other fans do.
how bad the reviews were because the album was too "mechanical" and was "missing the warmth of House"
FWIW, I wouldn't call myself a particular fan of house or dance music in general. This mechanicalness and lack of warmth is probably what I like about the album.
I really, really like Human After All. It’s probably their second most cohesive album, given how much they used the DigiTech Talker and DigiTech Synth Wah all over it. (Second to Random Access Memories in terms of being cohesive.)
Tron:Legacy still lives in my head. I'm still chasing the music I imagined after seeing the movie/hearing the soundtrack. Also half of Daft Punk Guy-Manuel de Homem-Christo's work in Kavinsky’s 'Nightcall'.
His name is Joseph Trapanese and I always wondered who worked with them on the album (for basically no credit) because, while I adore Daft Punk, very few pure dance music producers have the traditional-music/composer chops to pull off a film score.
I didn’t know until you pointed it out that he worked on Kosinski’s other film too. There are definitely some huge similarities, especially with the end title tracks.
Random Access Memories was such a letdown. It's an OK disco album that doesn't innovate at all. Masterful recording, it sounds great of course, but it didn't take any risks. I listened to it twice and that was it.
RAM brought back that funky sound which had been completely absent from contemporary dance music. It was a revival of old elements, mixed with modern electronic sound. A masterful mingling of elements of the past and the future coming together to form something not quite as good as either, but listenable and danceable, and most importantly impactful, because after that album there was a wave of artists emulating that sound. We wouldn't have gotten Uptown Special if we hadn't gotten Random Access Memories. That's my feeling on it anyway.
Exactly, fully agree.
If it would have been released as a Pharrell Williams Album "produced by Daft Punk", it woul have been met with different expectations.
And especially now it would perfectly blend with other Pharrell Williams Albums (i.e. "Happy" could have been on "Random Access Memories" without standing out at all...)
Guess you didn't listen to it on shuffle. There are multiple stories to be discovered depending on the order of tracks, like the one about the alien robot that crash lands on earth and strives to experience humanity. The album is RANDOM ACCESS Memories after all.
Interstella 5555 is still one of my favorite movies. It's an anime movie where the entire soundtrack is the album Discovery. There are no vocals, the entire story is told by only the animation and music, and it works incredibly well.
To be fair, they've done a lot more than just studio albums, from collaborations to live albums, concerts and whatever more. Personal favorite is probably Alive 2007 that went on repeat until Mom complained about the windows in the living room downstairs almost breaking.
Funny because Mom alternated between a Bob Marley greatest hits and Daft Punk Discovery and Human after all records when doing the ironing and cleaning at home.
No idea what she is listening to these days when doing those chores.
Interesting watch. It leaves some things unexplained though. How did Prydz end up as the artist on a song not in his own style, nearly identical to two previous artists making the same song, based on a re-recording of a sample? I wonder if the label was simply looking for a competent producer to make a version of what they thought would be a hit after having figured out how to clear the sample.
I mean, they’re retired, and I’m not holding out hope. But it’s remarkable that 30 years after their debut Daft Punk could put out an album and probably be as innovative and relevant as ever. That’s such a rare quality.
There’s rumours they’re working on more material. I wouldn’t blame them for telling the world they’ve broken up, just to take the pressure off. I can live in hope, right?
Very cool OP listening to the original samples compared against the different harmonizers and vocoders.
The Sennheiser VSM201 sounds so clean, I really like the analogue sound. The TC Helicon Talkbox Synth also sounds nice.
For the harmonizers, the Digitech Studio Vocalist EX sounds the best to me, but I also like the Korg ih Interactive Vocal Harmony for its spacey vocal effects.
I was a bit surprised by this article as it actually contradicts the account I always heard, which was their main vocal effect was a Roland VP-9000. If you listen to e.g. Harder Faster the effect is somewhere in between a vocoder & autotune, so I assumed that was the VP-9000. That said, this guy has clearly done his homework (pun intended) so I'm inclined to accept his version of events.
Yep, I’ve read that as well. The workflow and sound of the VP-9000 make me think it’s not what they used. It’s just so cumbersome to use, and it sounds pretty bad, and to my ears doesn’t match the results. The covers using DigiTech Vocalist EX models match so, so closely to the original songs.
It’s hard to know without an official word from our favourite robots, but I think you’d struggle to get the sound of any Daft Punk song out of a VP-9000. I’m really not sure where that rumour started.
I almost bought a VP-9000 to test, but sonically it’s so far off, I didn’t bother.
If you have a vocoder, running a drum machine through the modulator won't sound all that much like daft punk, but will probably sound familiar. And maybe become part of your sound.
If you don't have a vocoder, Behringer recently released one as a Eurorack module for $99. It's fine.
I went down a similar rabbit hole at the start of my PhD and I wish I’d written more of it up. One of my theories is that they combined effects quite often. For example, “harder better faster stronger” seems more likely to be a talk box recorded for a single note, then looped, then run through an AutoTune rack unit with MIDI inputs to repitch it. I mention this a little bit in a talk I have at ADC 2022 https://youtu.be/uX-FVtQT0PQ?feature=shared
Thanks for the talk link! I’m going to try writing some harmonizer code next, so your video is right up my alley. I believe IVL’s algorithm also isn’t FFT-based. That makes sense, given the CPU power around on consumer tech at the time.
As for Harder, Better, Faster, Stronger: It’s difficult to know for sure without comments from Daft Punk themselves, but the DigiTech Talker has such a unique, throaty sound, and it’s all over the Human After All album. My confidence varies with my guesses, but Harder, Better, Faster, Stronger is one of the more confident ones, given how distinctive the Talker is. They also used so much DigiTech gear, especially on that album.
Hopefully they’ll see the article and let me know which bits are wrong.
Ah, the Sennheiser VSM201. Just a $30K vocoder. Seems like it was $25K when it released in 1977, but also didn't get to sell even 50 units, so quite rare.
I guess you can get similar results with cheaper hardware, but if you have money and you have it around... ¯\_(ツ)_/¯
I didn't know the device. Also I didn't know that Kai Krause who later got famous through his Kai's Power Tools was an electronic music expert who sort of did sales for Sennheiser in 1977, according to this page (https://de.wikipedia.org/wiki/Sennheiser_VSM_201 - only on the German WP, it seems). He also wrote the manual for it.
I was surprised about how much better the Sennheiser sounds compared to the others. From the audio comparison in the first YouTube video. I did expect minor variations in the harmonics but the differences are quite significant between the models.
The other vocoder that sound almost as good is quite new and it seems to still be a prototype with a "contact us" price.
Yep! The Sennheiser and Ultimate VoIS are in their own league. There are some other rare high-end analogue vocoders that I would have loved to include in the comparison, but I don’t know anyone who owns them. The EMS vocoders are supposed to be amazing, too.
I can’t speak on Dromedary Modular’s behalf and I think rising parts costs have been an issue, but buying an Ultimate VoIS should be a fair bit cheaper than the Moog vocoder.
They used the VSM-201 for "Random Access Memories", their last Studio Album. At that time they didn't need such help anymore, they famously rented a huge amount of equipment and large studio-floors (i.e. they recorded simultaneously with microphones from different decades because Thomas heard a difference and wanted this reflected on the record).
Above all, the biggest help from his father was probably to insist that they keep the ownership of their music when signing with any label, regardless of any money, because as a producer he knew that this is how artists get screwed by record labels.
Yeah, owning their masters was a very smart move. I was surprised they rented the Sennheiser instead of buying one. Having said that, they don’t come up for sale very often.
Never had the pleasure of a Sennheiser but when working in radio I got
my hands on a lot of rack vocoders for doing branding, stings and
idents. Funny how the number 9000 comes up a lot, like Roland VP9000
and Eventide H9000. 80 and 90s vintage ones like Korg VC-10 or
Elektronik EM-26 had unique sounds, but tbh the modern digital
recreations are amazing models. There's not a world of difference
between vocoding, autotune, shifting, harmonising etc once you realise
how all the fx are now based in FFT, convolution etc - just different
variations on processing and control graphs - and so it's fun to
create your own vocal effects in things like
Max/MSP/PureData. Technically there's a distinction between "effects"
and "processing" in terms of how much of the direct (parallel) signal
is put through. Chers Believe is a yardstick for "effect", whereas a
lot of what I hear with Daft Punk (and Air, Kraftwerk) is quite
heavily processed as to disguise the original voice entirely - just
letting a bit of top/sibilant through to define the stops and
fricatives.
Analog vocoders are only nominally like analog FFTs.
The shape of the filters, the smoothing between the filters and the synthesis section, and (on some models) the patchability all create a very different result.
The reason the best analog vocoders are so expensive is because the filter for each band is much more complex than a plain old bandpass filter, with a much higher component count. Typically there's a flatter passband and a steeper slope than you'd expect.
You can do digital convolution with thousands of bins and it sounds nothing like analog vocoding. It's much cleaner, doesn't have those lovely harmonically spaced filter resonances, and creates sounds that can feel more acoustic than electronic.
Did you listen to the example audio in the video? Soft synths and digital emulation can be absolutely amazing these days, but the VSM201 and Ultimate VoIS are in their own league. It’d be pretty easy to pick them out from a blind test with other vocoders.
Oh, it also might be of interest that the IVL algorithm isn’t FFT-based. I think their harmonizers sound better than the rest, so maybe FFT isn’t the best way to go.
Yes exactly, I was really excited when I found out that you do not need a FFT to do speech processing.
If you look at the code of (phone/voice) codecs GSM/Speex/Opus you can see that you can estimate the spectral envelope (or the configuration of a physical tube model for the vocal tract) in time domain with linear prediction coefficients (LPC).
And it is simple, e.g. the often used Levinson-Durbin algorithm is just 22 lines of C code.
It is an interesting exercise to build your own vocoder from scratch that fits in a single screen page.
Many of the code snippets I have seen (which likely have already processed your voice) are just translations of the Fortran code of the book "Linear Prediction of Speech" by Markel and Gray (1976).
Ah yes, ladder or lattice filters. If you don't mind old fashioned
mailing lists there's still a few of hanging around in
MUSIC-DSP@LISTS.COLUMBIA.EDU where code gets shared.
Yeah, there’s a few, and I probably should have included more plugins. I don’t mind the built in Ableton vocoder. The built in Logic Pro one is a bit average. I like the XILS 201 plugin, but it’s not as nice as many of the hardware vocoders I tested. I guess it depends what you‘re after and how much vocoder you need.
(Note that the XILS 201 plugin requires iLok. I think iLok sucks.)
Thanks! I’ll give it a listen. Number of bands doesn’t necessarily dictate the quality of results — the Electro-Harmonix V256 has 256 bands, and it’s not even in the top 5 out of the vocoders I tested, imo.
Definately, Vocodex is probably one of the few vocoders you can turn to if you want something specifically uncharacteristic of a classic vocoder sound.
I feel like that’s a theme for Arturia’s plugins. I have their 1176 compressor and some other stuff. They’re nice, but use tons of CPU and can be a bit flaky at times.
Marc added some extra flavor https://mastodon.social/@marcedwards/114454783708869207
> This article is the longest piece I’ve published on Bjango’s site, and it took a couple of years of research. I purchased around 25 pieces of music gear. I emailed Imogen Heap, and to my surprise, someone from her team got back to me and confirmed the exact harmonizer used on Hide and Seek.
> It’s been a huge effort, and I’m confident it contains a lot of information that is not widely known. For those of you who are into Daft Punk, I hope it’s interesting.
It is unreal to me the amount of impact Daft Punk had with only four studio albums.
(They also did the Soundtrack for Tron:Legacy, basically their fifth studio album)
Anyway, I keep remembering how panned 'Human After All' was, and how bad the reviews were because the album was too "mechanical" and was "missing the warmth of House", while this is EXACTLY how the genre evolved in the years to come and none of those music experts saw this.
Many journalists did a retrospective of it a few years later and admitted that they misjudged it.
It's not that Daft Punk drove the industry in this direction, the album wasn't well-received by most at that time. They showed the destination of a journey while people didn't even realize they are traveling...
In the end, it appears that 'Random Access Memories' is one of their least innovative and "lasting" albums. It's probably their most successful one, the most complex to conceptualize and produce, but IMO it has the least unique character of all their productions.
Looking at the whole picture, the product of "Random Access Memories" is less the music, but the duo celebrating the process of production itself...
The reason that I dislike Human After All is that, quite simply, it's not fun.
It's dour. It's depressing. And it's repetitive in a way that feels tiresome; You can't dance to most of it or even really tap your foot to it.
What happened was that Daft Punk challenged themselves to make an album in six weeks and ended up with a showcase for a few neat guitar pedals and two fun songs (Robot Rock and Technologic), one of which consists almost entirely of a barely-changed sample (https://www.youtube.com/watch?v=PFwGQAEYqHs).
Electronic music certainly evolved in ways that made it less warm, but that didn't make it less fun. Case in point, Skrillex's early music is weird and playful despite relying on 'cold' synth sounds. Human After All is just cold.
>And it's repetitive in a way that feels tiresome; You can't dance to most of it or even really tap your foot to it.
Being robotic and repeatitive never stopped people dancing like crazy to techno all night. In fact, that's its very allure.
"in a way that feels tiresome" is the operative phrase here. I like the throwaway tracks on Homework that are twice as long and half as interesting more than I like most of Human After All. They're fun.
Not to be that guy, but techno is only repetitive if you are a) listening to boring/uninspired techno (the same kind of complaint being leveled here against HAA), or b) not paying attention.
Fine not to like it for those reasons. But maybe try listen to it next time from a more psych, garage, or post-rock perspective. They were influenced very early on by groups like Spacemen 3, and I see HAA as them doing something big and dirgey in this kind of school. I even recall the liner notes saying only "all guitars by Daft Punk".
The sound design is certainly interesting from the perspective of psych, garage, post-rock, etc., but the sound design doesn't make up for the songwriting. You need both. Or at least, I do.
I think it's a just a more experimental record. The lack of songwriting is kind of the thesis of the entire album, right? (technology, repetition, etc) It's aims are not to be Digital Love or their more conventional song-based stuff. If anything, it's almost a return to their origins like some of the more techno-orientated tracks on Homework.
That's what I meant with what it has in common with garage and psych. Not the textures or sounds, but the relentlessly experimental approach to convention. Not saying it is perfectly executes, but I think this what it is meant to be.
What you’re describing is an album that’s more fun to think about than it is to listen to!
It definitely has a theme, but unfortunately as executed it’s basically “what if we defied convention by using live instruments and making a shitty album?”
When you bring up psychedelic, experimental post-rock, I want to mention that one of my favorite bands is Battles. They’re a good example of what you can do when you defy convention!
Ha. Fair, but music that makes us think more than feel isn't necessarily always a terrible thing. John Cage, Steve Reich, even Battles is pretty intellectual in a way (it's quite literally math rock lol).
I guess we like different artists for different reasons, and you like Daft Punk in boogie mode. Which I definitely get, as it is their strong suit. But I just think this is album is a bit of an anomaly in their catalogue and might require having to unlearn them a bit to dig it.
Their live album Alive 1997 would have to be my favourite of theirs btw. But I don't hate Human After All as many other fans do.
how bad the reviews were because the album was too "mechanical" and was "missing the warmth of House"
FWIW, I wouldn't call myself a particular fan of house or dance music in general. This mechanicalness and lack of warmth is probably what I like about the album.
I really, really like Human After All. It’s probably their second most cohesive album, given how much they used the DigiTech Talker and DigiTech Synth Wah all over it. (Second to Random Access Memories in terms of being cohesive.)
Tron:Legacy still lives in my head. I'm still chasing the music I imagined after seeing the movie/hearing the soundtrack. Also half of Daft Punk Guy-Manuel de Homem-Christo's work in Kavinsky’s 'Nightcall'.
The composer who assisted them also worked with M83 for the Oblivion soundtrack, which I think is also pretty spectacular.
The Oblivion soundtrack is amazing! Love it. I’ve listened to it so many times. I haven’t seen the movie yet. I should do that.
Ah. Now it starts to make more sense.
Yes, both soundtracks are amazing. But I wonder how amazing they'd have been without his input.
His name is Joseph Trapanese and I always wondered who worked with them on the album (for basically no credit) because, while I adore Daft Punk, very few pure dance music producers have the traditional-music/composer chops to pull off a film score.
I didn’t know until you pointed it out that he worked on Kosinski’s other film too. There are definitely some huge similarities, especially with the end title tracks.
You may already know this, but Junkie XL/Tom Holkenborg has done some great film soundtracks, too.
Check out Tron: Legacy Reconfigured if you haven't (you probably have).
Random Access Memories was such a letdown. It's an OK disco album that doesn't innovate at all. Masterful recording, it sounds great of course, but it didn't take any risks. I listened to it twice and that was it.
RAM brought back that funky sound which had been completely absent from contemporary dance music. It was a revival of old elements, mixed with modern electronic sound. A masterful mingling of elements of the past and the future coming together to form something not quite as good as either, but listenable and danceable, and most importantly impactful, because after that album there was a wave of artists emulating that sound. We wouldn't have gotten Uptown Special if we hadn't gotten Random Access Memories. That's my feeling on it anyway.
Exactly, fully agree. If it would have been released as a Pharrell Williams Album "produced by Daft Punk", it woul have been met with different expectations.
And especially now it would perfectly blend with other Pharrell Williams Albums (i.e. "Happy" could have been on "Random Access Memories" without standing out at all...)
Guess you didn't listen to it on shuffle. There are multiple stories to be discovered depending on the order of tracks, like the one about the alien robot that crash lands on earth and strives to experience humanity. The album is RANDOM ACCESS Memories after all.
The stories are maybe 1/10000th the importance of the quality of the music, which is simply Ok.
Interstella 5555 is still one of my favorite movies. It's an anime movie where the entire soundtrack is the album Discovery. There are no vocals, the entire story is told by only the animation and music, and it works incredibly well.
I don’t even know what it is, haha (maybe it is an… Album Video). But it is great.
I guess there is the concept of a Rock Opera, but that doesn’t seem to have expanded much across the genres.
They were the official music videos for the album, looks like Daft Punk released all of them on YouTube.
This Playlist has them all in order: https://youtube.com/playlist?list=PLSdoVPM5WnndLX6Ngmb8wktMF...
I meant, I’m familiar with the thing, it just seems quite novel to have one continuous set of music videos that form a movie.
Sorry, in retrospect my comment was a bit confusing.
To be fair, they've done a lot more than just studio albums, from collaborations to live albums, concerts and whatever more. Personal favorite is probably Alive 2007 that went on repeat until Mom complained about the windows in the living room downstairs almost breaking.
The TRON:Legacy soundtrack was an incredible piece of work. Fit the movie exactly.
I have said many times that while TRON:Legacy wasn't a very good movie, it was absolutely the best-ever Daft Punk video.
Funny because Mom alternated between a Bob Marley greatest hits and Daft Punk Discovery and Human after all records when doing the ironing and cleaning at home.
No idea what she is listening to these days when doing those chores.
Is your Mom one of my exes?
dad?
Because those albums redefined the music industry, they also created the base platform for all live EDM concert with Coachella in 2006.
Daft punk had a large impact on music overall, they were ( are? ) really really good musicians.
Was lucky enough to see them live at a small venue in Manchester in the 90s when I was at University. Epic times.
Awesome! That would have been so good. I saw them just once — the Alive 2007 tour. Was a great day.
They did so much more. Thomas made Call On Me, a track he found unsatisfying, and it still ended up as a massive hit because it was stolen from him.
Imagine inventing pumping sidechain compression and finding it unsatisfying.
A YouTube video (that the algorithm loved) about all this and is super interesting: https://youtu.be/wyYAiU4DKUY?si=CpcOuhzAvXs7fqWF
Interesting watch. It leaves some things unexplained though. How did Prydz end up as the artist on a song not in his own style, nearly identical to two previous artists making the same song, based on a re-recording of a sample? I wonder if the label was simply looking for a competent producer to make a version of what they thought would be a hit after having figured out how to clear the sample.
Both of them had separate labels and released a lot of other artists. See DJ Mehdis super hit for one of my favourite tracks ever :)
Specifically Roulé and Crydamoure (unsure if spelling correctly)
Absolutely. Hopefully there’s at least one more.
I mean, they’re retired, and I’m not holding out hope. But it’s remarkable that 30 years after their debut Daft Punk could put out an album and probably be as innovative and relevant as ever. That’s such a rare quality.
How can this be?
There’s rumours they’re working on more material. I wouldn’t blame them for telling the world they’ve broken up, just to take the pressure off. I can live in hope, right?
At least if they don't produce another album, we still get to see them perform at Burning Man this year, out by the trash fence
Very cool OP listening to the original samples compared against the different harmonizers and vocoders.
The Sennheiser VSM201 sounds so clean, I really like the analogue sound. The TC Helicon Talkbox Synth also sounds nice.
For the harmonizers, the Digitech Studio Vocalist EX sounds the best to me, but I also like the Korg ih Interactive Vocal Harmony for its spacey vocal effects.
This is a really great deep dive, I wish I could upvote more to reward this kind of quality work.
Thank you! Your kind comment helps. :)
I was a bit surprised by this article as it actually contradicts the account I always heard, which was their main vocal effect was a Roland VP-9000. If you listen to e.g. Harder Faster the effect is somewhere in between a vocoder & autotune, so I assumed that was the VP-9000. That said, this guy has clearly done his homework (pun intended) so I'm inclined to accept his version of events.
Yep, I’ve read that as well. The workflow and sound of the VP-9000 make me think it’s not what they used. It’s just so cumbersome to use, and it sounds pretty bad, and to my ears doesn’t match the results. The covers using DigiTech Vocalist EX models match so, so closely to the original songs.
It’s hard to know without an official word from our favourite robots, but I think you’d struggle to get the sound of any Daft Punk song out of a VP-9000. I’m really not sure where that rumour started.
I almost bought a VP-9000 to test, but sonically it’s so far off, I didn’t bother.
Reading the title I thought this was about extraordinary singing techniques. But nice article anyway.
Obligatory post of the legendary Bode Vocoder demonstration: https://www.youtube.com/watch?v=5kc-bhOOLxE
Yes! Not sure how many times I’ve watched that, but I will watch it at least once every time someone posts it.
Outstanding article, don't skip the youtube videos!
Thank you!
If you have a vocoder, running a drum machine through the modulator won't sound all that much like daft punk, but will probably sound familiar. And maybe become part of your sound.
If you don't have a vocoder, Behringer recently released one as a Eurorack module for $99. It's fine.
Did you buy one? I think it’s very likely I’ll get a VC16, but they’re not in stock anywhere I purchase from yet.
Yes. Still in stock on eBay.
https://www.ebay.com/itm/335903639694
Ah, cheers! I’m in Australia, and it seems like Alto won’t ship here. I’m sure they’ll appear on a store that does ship to me soon.
I went down a similar rabbit hole at the start of my PhD and I wish I’d written more of it up. One of my theories is that they combined effects quite often. For example, “harder better faster stronger” seems more likely to be a talk box recorded for a single note, then looped, then run through an AutoTune rack unit with MIDI inputs to repitch it. I mention this a little bit in a talk I have at ADC 2022 https://youtu.be/uX-FVtQT0PQ?feature=shared
Thanks for the talk link! I’m going to try writing some harmonizer code next, so your video is right up my alley. I believe IVL’s algorithm also isn’t FFT-based. That makes sense, given the CPU power around on consumer tech at the time.
As for Harder, Better, Faster, Stronger: It’s difficult to know for sure without comments from Daft Punk themselves, but the DigiTech Talker has such a unique, throaty sound, and it’s all over the Human After All album. My confidence varies with my guesses, but Harder, Better, Faster, Stronger is one of the more confident ones, given how distinctive the Talker is. They also used so much DigiTech gear, especially on that album.
Hopefully they’ll see the article and let me know which bits are wrong.
Ah, the Sennheiser VSM201. Just a $30K vocoder. Seems like it was $25K when it released in 1977, but also didn't get to sell even 50 units, so quite rare.
I guess you can get similar results with cheaper hardware, but if you have money and you have it around... ¯\_(ツ)_/¯
I didn't know the device. Also I didn't know that Kai Krause who later got famous through his Kai's Power Tools was an electronic music expert who sort of did sales for Sennheiser in 1977, according to this page (https://de.wikipedia.org/wiki/Sennheiser_VSM_201 - only on the German WP, it seems). He also wrote the manual for it.
His German WP page also claims that he sold a VSM 201 to Neil Young in 1982! https://de.wikipedia.org/wiki/Kai_Krause
English WP has less details on that part of his life, especially the VSM 201 :) https://en.wikipedia.org/wiki/Kai_Krause
Kai Krause wrote the VSM201 manual and helped sell it at Sennheiser?! You’re blowing my mind. I didn’t expect that info today.
> His German WP page also claims that he sold a VSM 201 to Neil Young in 1982!
Just in time for https://en.wikipedia.org/wiki/Trans_(album) !!!
Wow, six of nine tracks with the vocoder. I guess the 25k USD machine had to pay off :-)
Ha! Now that's unexpected!
The article says that Daft Punk rented it.
I was surprised about how much better the Sennheiser sounds compared to the others. From the audio comparison in the first YouTube video. I did expect minor variations in the harmonics but the differences are quite significant between the models.
The other vocoder that sound almost as good is quite new and it seems to still be a prototype with a "contact us" price.
Yep! The Sennheiser and Ultimate VoIS are in their own league. There are some other rare high-end analogue vocoders that I would have loved to include in the comparison, but I don’t know anyone who owns them. The EMS vocoders are supposed to be amazing, too.
I can’t speak on Dromedary Modular’s behalf and I think rising parts costs have been an issue, but buying an Ultimate VoIS should be a fair bit cheaper than the Moog vocoder.
Thomas Bangalter's father was a French disco producer and songwriter who helped them a lot in their early career. Sure helps.
They used the VSM-201 for "Random Access Memories", their last Studio Album. At that time they didn't need such help anymore, they famously rented a huge amount of equipment and large studio-floors (i.e. they recorded simultaneously with microphones from different decades because Thomas heard a difference and wanted this reflected on the record).
Above all, the biggest help from his father was probably to insist that they keep the ownership of their music when signing with any label, regardless of any money, because as a producer he knew that this is how artists get screwed by record labels.
Yeah, owning their masters was a very smart move. I was surprised they rented the Sennheiser instead of buying one. Having said that, they don’t come up for sale very often.
I wonder how many are still in working condition today? Can’t be many. I’d love to see one in person, one day.
It's Daft Punk, are we really surpised? :-P
Never had the pleasure of a Sennheiser but when working in radio I got my hands on a lot of rack vocoders for doing branding, stings and idents. Funny how the number 9000 comes up a lot, like Roland VP9000 and Eventide H9000. 80 and 90s vintage ones like Korg VC-10 or Elektronik EM-26 had unique sounds, but tbh the modern digital recreations are amazing models. There's not a world of difference between vocoding, autotune, shifting, harmonising etc once you realise how all the fx are now based in FFT, convolution etc - just different variations on processing and control graphs - and so it's fun to create your own vocal effects in things like Max/MSP/PureData. Technically there's a distinction between "effects" and "processing" in terms of how much of the direct (parallel) signal is put through. Chers Believe is a yardstick for "effect", whereas a lot of what I hear with Daft Punk (and Air, Kraftwerk) is quite heavily processed as to disguise the original voice entirely - just letting a bit of top/sibilant through to define the stops and fricatives.
Analog vocoders are only nominally like analog FFTs.
The shape of the filters, the smoothing between the filters and the synthesis section, and (on some models) the patchability all create a very different result.
The reason the best analog vocoders are so expensive is because the filter for each band is much more complex than a plain old bandpass filter, with a much higher component count. Typically there's a flatter passband and a steeper slope than you'd expect.
You can do digital convolution with thousands of bins and it sounds nothing like analog vocoding. It's much cleaner, doesn't have those lovely harmonically spaced filter resonances, and creates sounds that can feel more acoustic than electronic.
Did you listen to the example audio in the video? Soft synths and digital emulation can be absolutely amazing these days, but the VSM201 and Ultimate VoIS are in their own league. It’d be pretty easy to pick them out from a blind test with other vocoders.
Oh, it also might be of interest that the IVL algorithm isn’t FFT-based. I think their harmonizers sound better than the rest, so maybe FFT isn’t the best way to go.
Yes exactly, I was really excited when I found out that you do not need a FFT to do speech processing.
If you look at the code of (phone/voice) codecs GSM/Speex/Opus you can see that you can estimate the spectral envelope (or the configuration of a physical tube model for the vocal tract) in time domain with linear prediction coefficients (LPC).
And it is simple, e.g. the often used Levinson-Durbin algorithm is just 22 lines of C code. It is an interesting exercise to build your own vocoder from scratch that fits in a single screen page.
Many of the code snippets I have seen (which likely have already processed your voice) are just translations of the Fortran code of the book "Linear Prediction of Speech" by Markel and Gray (1976).
Ah yes, ladder or lattice filters. If you don't mind old fashioned mailing lists there's still a few of hanging around in MUSIC-DSP@LISTS.COLUMBIA.EDU where code gets shared.
I thought it was phase synchronous overlap add, but I just checked and now I'm not so sure.
Has anyone got more details?
That gives me something to research! I’ve only scratched the surface of IVL’s algorithm, but intend to look into it further.
I love, no luuuurrve, this article. Just fantastic research and fantastically useful for a music project I'm workin' on.
Tangentially related; here's a great example of the classic daft punk vocal synth: https://www.youtube.com/watch?v=0mXLNnZvGJw
But... Skala?!
We’re working hard on it! It’s going to be good. There’s some more info on this page: https://bjango.com/mac/skala/
Thanks. :)
Any good software vocoders out there?
Yeah, there’s a few, and I probably should have included more plugins. I don’t mind the built in Ableton vocoder. The built in Logic Pro one is a bit average. I like the XILS 201 plugin, but it’s not as nice as many of the hardware vocoders I tested. I guess it depends what you‘re after and how much vocoder you need.
(Note that the XILS 201 plugin requires iLok. I think iLok sucks.)
Fl studios "vocodex" is likely the most advanced vocoder there is.
128 bands, band redistribution, ability to further route using patcher, just look at its manual to see :)
Though its shrouded by the standard "FL is for rap dumb dumb" stigma. Marketing is one hell of a drug.
Thanks! I’ll give it a listen. Number of bands doesn’t necessarily dictate the quality of results — the Electro-Harmonix V256 has 256 bands, and it’s not even in the top 5 out of the vocoders I tested, imo.
Definately, Vocodex is probably one of the few vocoders you can turn to if you want something specifically uncharacteristic of a classic vocoder sound.
The built-in one in ableton is fine by me
I really like Arturia's though it is a bit finicky, a CPU hog, and takes some post processing to clean up.
I feel like that’s a theme for Arturia’s plugins. I have their 1176 compressor and some other stuff. They’re nice, but use tons of CPU and can be a bit flaky at times.
For stuff that doesn't need sidechain I run it in AudioGridder and that resolves the issue but the vocoder needs sidechain.
Tangentially related if you make music in the box and want some simple Daft Punk breakdowns to experiment from:
https://reverbmachine.com/blog/daft-punk-homework-synth-soun... https://reverbmachine.com/blog/daft-punk-discovery-synth-sou...