There is an idea that a lot of classical players carry around, that the most important part of a note is the part after you stop playing it, the way the sound hangs in the hall and dies away, and I think about that idea a lot in our edit bay, because it describes exactly what most podcast editing software destroys. Noise reduction treats that decay as hiss, a compressor squashes the quiet ending up to the same level as the attack, and a tool that cuts on silence chops the reverb tail off before it finishes, so the music on a typical edited podcast sounds like it was recorded in a cupboard.
For most shows nobody notices, right, because most shows are two people talking. For a classical music podcast in Bloomington, it is the first thing your audience will notice, because your audience is full of people who spend hours a day listening to exactly that decay. Indiana University is in Bloomington and IU's Jacobs School of Music is one of the largest music schools in the country, which means the town has an unusual density of performers, faculty, graduate students and private teachers, and a good number of them are now recording conversations about repertoire, practice, auditions and teaching. This post is about how we edit those shows at Pixel Samy Podcast, from Dubai, and what a careful edit needs to protect.
Why is a classical music podcast harder to edit than a talk show?
Because it contains two kinds of audio that want opposite treatment, speech that should be levelled and cleaned so it is easy to listen to, and performance that should be left almost untouched so its dynamics and room sound survive.
On a talk show we smooth the levels, so a host who leans back from the microphone does not suddenly disappear, and we clean up hum and room noise. If you apply that same chain to a passage of Brahms, right, you have just taken away the pianissimo the performer worked a month to control. So the first thing we do on a classical show is separate the two, treating the conversation as speech and treating every performance excerpt as its own piece of audio with a much lighter hand, matched in overall loudness to the talking around it so the listener is not reaching for the volume, but otherwise allowed to breathe.
There are published targets for the speech side. Apple, for instance, gives a loudness recommendation for podcasts in its audio requirements for creators, and we master the full episode to sit comfortably in that range. The judgment comes in at the joins, where a quiet chamber passage meets a loud laugh from the host, and that is not something a preset can decide for you. If you want the detail of how we approach cleanup on spoken audio generally, the page on podcast audio cleanup that listeners never notice covers it, and everything on that page applies to your conversation, just not to your playing.
You can probably already hear the difference in your head, the excerpt that sounds like a concert and the one that sounds like a voicemail.
How should performance excerpts be handled in the edit?
Each excerpt should start and end on a musical boundary rather than wherever the conversation happened to pause, with the room sound left in and the levels matched by ear, so the listener hears the phrase the performer intended.
That sounds obvious, right, but it is where most generic editing goes wrong. A host says "let me play you the opening", plays eleven bars, stops mid-phrase to make a point, and an editor who is not listening musically will cut on the stop. We would rather extend to the end of the phrase, or trim back to the cadence before it, and if the excerpt was played on a different microphone from the conversation, we match the tone so the switch does not jar.
Who owns the music you play on your show?
That depends on who wrote the piece, who made the recording and who you recorded it for, and since we are editors and not lawyers, the honest answer is that you need to confirm it before the episode goes out, and we will build the edit around whatever you tell us is cleared.
Here is the general shape of it, as information and not advice. A composition can be old enough to be out of copyright while a particular recording of it is very much owned by someone, a label, an orchestra, a festival, sometimes a university. Playing a commercial recording on your show is a different situation from playing your own performance, and your own performance recorded at a school recital may still have its own rules attached. Music teachers and students in particular should check, because recordings made in an institutional setting are not always the performer's to publish.
What we do in the edit is simple, right, and it does not change from show to show. You tell us which excerpts you are cleared to use, and we use those. If something in the raw files sounds like a commercial recording you did not mention, we ask before it goes in. We never drop music into your episode that you have not supplied or approved. And if a passage needs to come out after a rights holder gets back to you, we cut it and re-export. To be very honest, this is the conversation most music hosts skip until a video gets flagged, and it is much cheaper to have it at the start.
Do I need a Bloomington editor who can sit in on the recording?
You need someone in the room for recording a performance well, a good engineer, decent microphones and a quiet space, but the editing that follows does not need to happen in Bloomington at all, because it happens in the files.
If anything, this is the one show type where I would push you to spend on the recording before you spend on anything else. A practice room with hard walls and a laptop microphone cannot be fixed later, and no editor anywhere, local or remote, can put the hall back into a dry recording. Once you have good files, though, right, distance stops mattering. I wrote about this more broadly in the post on why distance matters less than people think when choosing an editor, and the short version for a musician is that the person who records you and the people who edit you do not need to be the same, or in the same country.
For the editing side of music shows specifically, our page on podcast editing for musicians who actually play on the show goes through formats like live sessions and multi-camera performance, so I will not repeat that here. And if you are in another part of the state with a very different kind of show, the Columbus, Indiana post on engine makers and industrial technology shows is a good reminder of how differently two towns an hour apart can need to be edited.
How do files get from Bloomington to Dubai, and how long does it take?
Bloomington is on Eastern Time and Dubai is 9 hours ahead of you in winter and 8 hours ahead during US daylight saving, so if you upload after a 7 pm rehearsal, the files reach us at 4 am Dubai time in winter, or 3 am in summer.
In practice that means, right, you finish recording, you drop the camera files and the separate audio tracks into a shared folder, and you go to bed, while the edit is already underway on our side. Big multi-track recordings from a recital are fine, there is no cap on the size of what you upload, and we would always rather have every isolated microphone than a single mixed file, because separate tracks are what let us treat the speech and the music differently. The remote podcast editing page goes through the whole cross time zone setup, and for music shows the one thing I would add is to label your excerpts in the folder, even roughly, "Bach, second movement, take two", because it saves a round of questions.
Your first cut arrives two days after the upload. If it is a long masterclass with a lot of performance in it, it can take a day more, and we tell you as soon as we have heard the files. You send one consolidated set of notes, timestamps are perfect, and the fixes reach you inside a working day.
What does this cost for a performer, a teacher or a student?
For a show recorded on one camera, our Starter plan is 260 dollars an episode for up to 90 minutes, which suits the long, unhurried conversations musicians tend to have, and a shorter episode costs less.
If you record a teaching show with a second camera on the hands or the bow arm, that moves you to the multi-camera plan, and the numbers for every length are in the pricing section. If your show is audio only, which plenty of music podcasts are, we edit those too and price them on a short call once we know how much performance is in each episode. Students on a budget, right, can pay per episode and stop whenever the semester gets heavy, there is no contract to keep paying through exam season.
Teachers building a show as part of their studio, for instance a series that answers the questions your students ask every week, might also like the post on podcast editing for teachers and educators, which looks at structuring a show around lessons rather than interviews.
When should a music host pick a different option?
When the episode is mostly a concert with a sentence or two between pieces, hire a classical recording engineer instead of us, because what you have is music production with some talking in it, and that calls for a different specialist than a podcast with some music in it.
Similarly, if you want someone to come to the hall, set up the microphones and run the session, you need a local engineer, and we are not that, we are a remote team of nearly fifty in Dubai who edit what you send. A freelancer who is also a trained musician can be a wonderful fit for a single show if you can find one with time, though it gets harder to rely on one person once you publish weekly. Where we make sense is the regular show, mostly conversation with performance woven in, where you want the talking cleaned up properly, the music left alone, and every episode delivered on a schedule you can plan a season around. You also keep every file we produce, including the project, so you are never locked to us.
I would rather leave a little room noise in a phrase than take away the moment the note dies.
What makes a classical show worth listening to twice?
Usually it is the moment the guest stops explaining and plays, and the listener hears the thing the whole conversation was about, so the edit has to make sure that moment sounds like music and not like a phone recording of music.
The talk around it matters, and we will make the talk easy to follow, but your audience came for the sound, right, and they will forgive a slow anecdote long before they forgive a ruined pianissimo. So the way I think about a Bloomington music show is very very simple, edit the conversation like a podcast and protect the performance like a recording, and keep the two from ever being treated the same way.
So yeah. That's my way of saying it.