top of page

Mixing

When it comes to translating the multitrack live recording to an energetic and clear mix, it is important to consider how you as the artist want to be represented. Both Kieran Bowdidge and Drew Bisset emphasised this when listening to performances they are mixing before making creative mix choices. There are many technical differences to consider when mixing a live gig recording compared to professional studio recording so I will go through some key differences and simple techniques to produce a mix that can maintain the energy from a performance while having the clarity of a professional, high quality studio recording from the different types of audio recordings I captured. 

Contents

Bleed

First and most importantly:  the key to mixing a live recording is working with the bleed and not against it. Everything will bleed into every microphone (especially those cymbals into every microphone on stage) so mixing a live recording is not as simple as a studio recording and making every microphone source perfect with excessive gating and EQ will ruin the audio quality and overall tonality of the recording. Drew Bisset explained that the bleed does blend elements together better and Kieran Bowdidge described it as a part of the sound, carrying the vibe. When assessing the performance for sections to use for social media and promotional content, keep in mind that any sort of pitch correction or quantization will sound awful. The wrong note or out of time snare will still be picked up in all of the other microphones, so pitch shifting or quantizing a source will not fix a sound, only ruin the quality of that channel and the bleed from everything else in that channel. Embrace the rawness of the live performance and those elements that make it live and real, but for social media and short form content, you have the luxury to choose the better chorus or song.  

IMG_0634.jpeg

Crowd Microphones and Recordings

Secondly, the point of the live gig recording at a venue is to capture the audience there and include that as part of the mix. With the crowd mics, subtractive EQ to roll off the low-end muddiness and high-end harshness were points made by both Kieran, Drew and Sam from Sound on Sound. Adding additional reverb will mould the space and enhance the illusion of being in the room with the crowd using a room or hall reverb setting. Keep the processing of the dedicated crowd simple, but “ride the fader” and automate the volume throughout the song or sections where crowd interaction or involvement is key to the vibe of the moment. Alternatively, if your set up doesn’t include dedicated crowd mics, use additional vocal mics or stage mics that are not actively capturing a different independent element and that are picking up the audience, boost their volume and have them panned left and right to emulate a crowd mic and allow for the crowd presence to be incorporated.

Automation and Fades

Continuing on the point of automating, the whole performance and recording will be dynamic. No one set volume for a channel will be perfect all the way through, so automating the volumes in points of impact is key to keep the flow of the audio and bring out the elements creating the energy in the right moments. Examples of key moments to consider volume automation include guitar solos, when a vocalist stops singing and the bleed into the vocal mic ruins the mix of the instrumental, or when the crowd participates more in the chorus of a popular song. It is also key to utilise fading in and out at the beginning and end of the audio you wish to use, whether that is a short, almost instantaneous fade or a slower gradual dramatic fade for effect. The prevents clicking when the waveform is cut off or when a different audio starts playing immediately. If editing audio sections to accompany a video edit, use cross fades to smoothly transition the audio from one clip to the next, once again, preventing audio clicks or a moment of silence gap between clips. 

Screenshot 2026-05-18 at 14.21.05.png

Compression

Compression is used to reduce and average out the dynamic range, by lowering the level of the signal that passes a certain loudness threshold. It reduces the difference in the loudest and quietest parts of the audio, usually by squashing the transients.  

​

There are many types of compressors that you will come across in DAWs; clean digital compressors, Voltage-Controlled Amplification (VCA) compressors, FET/1176 compressors, Optical Tube/LA-2A compressors, Glue compressors, which all have their distinguishing characteristics as to why we use them in mixing. The hardware emulations such as VCA, FET and Tube compressors are known for the harmonics and distortion they add to the audio signal, while modern stock digital compressors will remain clean and not add more character to your audio. They also have different capabilities in their natural ability to attenuate the signal, at difference attack and release speeds, knees and ratios. There are many elements to compressors and the different types and characteristics to explain them all but here are a few points that will be useful for mixing live audio to achieve clarity and energy.

 

Fast attack times will make the compressor hit the transients faster, but a slower attack will let the peak of the transient to hit after passing the loudness threshold before turning on and reducing the signal. The release time is the amount of time it take for the compressor to release and stop reducing the signal level. The knee is how wide or thin the threshold is, and when the compressor starts activating, either sharply as the threshold is hit or gradually as the level approaches the threshold. Finally, the ratio is basically the amount of gain reduction that happens. This diagram below explains clearly. 

 

Multi-band compression is a commonly used form of compression that compresses the audio differently in each frequency band, allowing for independent compression settings in each area, helping correct frequency-based issues alongside an EQ. Fab-Filter Pro-MB is an example of one (£150).

​

Side-chaining a compressor to a different source or compressor makes the compressor react according to the input source (trigger), but affect the audio channel it is inserted on (the target). Commonly, side-chain compression is used to ‘duck’ the bass line (reducing its volume momentarily) according to the transients of the kick drum to make room for the kick to punch through but can be used in many contexts. 

De-essing is a form of side-chain compression, using the frequency information in a signal to trigger a compressor (to act on that certain frequency, the whole signal or just the high end) when a specific frequency passes the defined threshold level in the signal. De-essing is very useful to use on harsh vocals, reducing sibilance around the 5-8kHz range, depending on the vocalist, but can be used on different elements, such as guitars to reduce high end and harsh harmonics. 

​

Parallel compression is common way to add more saturation and punch, but, maintain dynamics in a mix. You can add a 100% wet, heavy compressor (fast attack, slow release, low threshold, hard knee) to a bus/aux send and feed in the channel to it on a send to have the second (parallel) signal blended with the dry, unprocessed signal. 

 

You will commonly hear the term “bus compression” which is fundamentally a compressor used on the master bus (or just on a bus affecting an independent group of tracks also), usually a VCA compressor, to “glue” everything together. It ‘gently squeezes’ the mix, as BBC mixing engineer Sara Carter explains it, “like tidying a stack of papers before stapling them together”. It brings all the tracks together, smooths out the dynamics of the complete mix and unifies it, rather than just treating each individual track of audio. Most of the time, these are set with a low ratio, soft knee and slow attack and auto-release.

​

Compression is a very handy tool, and you will find it will be used a lot on a live recording alongside EQ. Using a processing chain of an EQ followed by a compressor followed by another EQ and compressor (and on to your heart’s content) is an effective method to balance out frequencies and dynamics without completely cutting out frequency ranges or squashing transients in recordings where bleed is present. Alternatively, mixing engineer Miles Davis advises compressing before EQing to control the dynamics of the environment and cause the EQ to affect the signal more consistently. Additionally, the nature of reducing the loudness after a threshold, will comparatively raise the level of the quieter parts, meaning more bleed will be present in the signal. This is when volume automation will be your best friend, between sections in the vocal mic or drum hits in the drum mics. One creative and useful way to use compression for a live recording by Miles Davies is to heavily compress the overhead if you want a more prominent cymbal sound, especially if there are no spot microphones on the hi-hat or ride cymbals that might be important to the mix. Here is a comparison of changing the processing chain for a song from Day We Ran's first show, using the stereo room recording, with the exact same EQ and compression settings, just in a different order. 

​

​

​

​

​

​

​​​​

 

 

Another rule of thumb is to make sure you can’t “hear” the compressor, if you hear the transients distorting or the release of the compressor losing the groove of the track, you will have either applied too much compression at once, or used the wrong type of compression than what is required for the sound. Firstly, try to reduce the threshold of the compressor if it is being hit hard, or lower the ratio of the compressor if the signal isn’t initially loud in the compressor but still distorting. If the groove is being lost, adjust the attack and release times; reduce the release time if it stays consistently compressed and loses dynamics, increase the attack time if too much transient is not being compressed, or reduce the attack time the transients are not present enough. 

Day We Ran 1 - EQ before Compression
Day We Ran 1 - Compression before EQ
16b7d60bcef8-image-2-audio-compressor-ratio.jpg

EQ

EQ is short for equalisation and is one of the most fundamental audio processing tools, used to control the level of certain frequencies or a range of frequencies, by cutting unwanted harmonics or frequencies and boosting the ones you do want, to create a more pleasant tonal balance.

Just like compression, EQ is a whole other world of theory, but I will cover the basics necessary to mix your live set. If you are new to mixing, I advise sticking to using a multi-band parametric EQ, the ones that are stock on most DAWs, that allow you to visually see frequency content, manually adjust the frequency, q and type of EQ shape. 

​

You will commonly use high pass and low pass filters on most tracks. A high pass filter is a filter that allows frequencies higher than a certain frequency ‘pass’ through, reducing the amount off low-end information. A low pass is the exact same but allows the frequencies that are lower than a certain frequency ‘pass’ through, reducing the amount of high-end information.

​

Making EQ choices comes down to the frequency content of the instrument and how that sits within the mix of all the elements together. Use subtractive EQ for frequencies and harmonics that are not pleasant within the channel to remove them, such as around 6-7kHz in vocals for sibilance, around 4kHz in guitars to let the vocals and snare sit in the mix louder, or to roll off bass in tracks where bass information is not needed or to counteract the proximity effect in microphones, especially vocals. Use additive EQ if you want to bring out areas that may be lacking in frequency content or where elements need to be more present in the mix, such as the click of a kick drum around 3-4kHz, vocal clarity at 4kHz or boost high end information, known as ‘air’ to guitars or vocals. Each instrument and vocalist will have different frequencies that are beneficial to the sonic information of the mix, and some unpleasant harmonics. Listening to the channel in solo is not useful when EQing to clean up the signal, always make EQ decisions in consideration of the whole mix and frequency content. It is common to attenuate frequencies in elements to “make room” for other elements, such as guitars in the high-mids so the vocals have space in the mix. EQing vocals and guitars to work together is a large part of mixing live indie rock gigs, when it is common to have multiple guitars and up to 3 male vocalists, and you must control the muddiness of all these mid-range elements. I will walk through my processes to achieve this is the multiple types of recordings. 

Audible-Genius-Eq-frequency-chart.jpg

Reverb

Reverb can be a handy tool to exaggerate the live room feeling, like the listener is still in the venue. Reverb (short for reverberation) is the reflections of sound within a space, which is emulated in music production for effect and make elements feel bigger and create a sense of space.

​

Using reverb on crowd and room mics can replicate the larger space that the gig was recorded in. It can also make the close mic’d or DI’d elements feel more live and blended into the space, like the snare drum or acoustic guitar. 

​

To use reverbs, it is standard to set up your reverb on an aux channel and use a send from the audio channel. This way many elements can be sent to the same reverb without overloading your computer’s processing. Make your reverb 100% wet on the aux and choose your level to be sent on the channel strip’s send. Also, a traditional reverb technique is to add an EQ before the reverb (on the aux channel’s insert) and roll off the high end after 10kHz and low end before 600Hz. This is called the Abbey Road technique and reduces muddiness in the bass and harsh washiness in the high end.

​

Once again, time-based processing like reverb should be used with caution to not affect the relationship and cohesion between all the tracks. Stick to using space emulation reverbs, such as halls, rather than springs or plates, and dial in the decay to replicate a believable space.

Stem Splitting

Stem splitting uses AI to extract different elements from a single audio source and create stems. Stem splitting can be a useful tool, however in the context of live recordings, I advise against relying on it. Artefacts are created from the artificial separation from one audio file, and it doesn’t create perfect separation, primarily with the harmonics of certain elements. I have experimented with Ableton Live 12’s stem separation device and here is a sample from Day We Ran's first show room stereo recording. 

​

The separation is quite impressive but as you will hear the tone and harmonics of each element are not perfect and sound quite empty on their own, trying to correct these in individual stems will cause gaps in your mix if you EQ out the guitar harmonics in the vocals or compress and volume automate elements for example. 

Screenshot 2026-05-11 at 21.06.21.png

Overdubbing 

One technique to correct or enhance recordings is to use samples, over dub or use recordings over different shows mixed together. For the purpose of short form content and promotion, you have the luxury of only picking usually 15-second, 30-second or 60-second sections to use for content, this can be chopped up bits of a song like the video created by Letters to Lions or picking the best chorus of the song. Even capturing the mess ups and awkward crowd moments can be super entertaining content for social media, embrace it, keep it real. A Waves Learn article, written by engineers at a industry leading plug-in and signal processing developers, suggest to use transient shaper plugins can re-shape transients on drum close mics rather than overdubbing, such as their plug-in Smack Attack ($34.99USD)

Master Bus

Once you are almost satisfied with the mix, and only once you have mixed to satisfactory level, should you touch the master bus. Using the master bus will bring everything together right at the end. On the master bus, apply a bus compressor, a tape saturator or emulator, EQ and limiter.

 

The bus compressor will glue the mix together, as I went through in more detail in the compression section of mixing.

A tape saturator will add more warmth and colour to the mix, with tape emulators such as the Universal Audio tape plug-ins providing a wide range of characteristics to play around with to best suit your mix. 

After this saturation, finalise an overall frequency balance with a gentle EQ in areas that need more emphasis in the overall mix.

Finally, you will want to apply a limiter to ensure the mix is loud enough to export and use for content.​

​

Once you have applied these elements and are happy with the final product, put a metering plugin at the end of your master bus chain. I recommend Youlean loudness meter plug-in as a free option! This is key to analyse the LUFS and true peak level of your audio, to ensure it is suitable for sharing. Audio that does not meet the requirements of digital platforms will be squashed with a compressor if it is too loud or dynamic or the gain will be increased if it is too quiet, reducing the quality of the audio with additional noise or distortion. 

​

LUFS (short for Loudness Units in Full Scale) is the loudness (in decibels) measurement based on the tonal balance of the sound and how our ears will perceive it. For the purpose of loudness mastering, the integrated LUFS reading is the important information, taking the average loudness of the audio.

The true peak (dBTP) meter will tell you what the peak level will be when the audio if played out of a digital-to-analogue converter (speakers or headphones etc). This is important to assess as DAWs use digital audio and have sample peak levels, but when this information is turned into analogue voltage it can be at converted at a higher loudness level. This meter attempts to reflect what the digital-to-analogue conversion true peak will be to avoid clipping and distorting the audio. 

 

For content platforms the ideal master volume levels are:

YouTube = -14LUFS 

Instagram and TikTok = between -10 and -12 LUFS

Facebook = -13LUFS 

Ensure there is a true peak value of -1 to -2dBTP to leave headroom and avoid analogue clipping. 

Day We Ran 2 - No Master Bus
Day We Ran 2 - Master Bus

Specific Mix Examples

DEVELOPED BY DUMASTAR

loading.gif
Day We Ran 1

Foaming was the last song of Day We Ran’s set, so I am using this example as it is high energy, and a crowd favourite with audience participation. I am working with the stereo of the room, so the mixing is basically a stereo master. My processing was a mix of EQ, Compression, and tape saturation. One technique I tried was to compress before using EQ. This dramatically improved the clarity of the stereo audio as the EQ felt more evenly applied. I had one 1176 emulation before applying EQ with an SSL bus compressor after compared to this version with both compressors afterwards. Both tracks have the same EQ and tape saturator plug-in to add more warmth. There are 2 7-band parametric EQs and 2 1-band parametric EQs all with small adjustments, especially many attenuations in the mid-range area as 2 guitars, a bass guitar and 3 male vocals were fighting for that space. 

Betty Taylor

This set was recorded without any room mics, only a close mic multi-track and LR crowd mics from the stage. To emphasise the live recording energy, a room reverb was added to channels such as the vocals, guitars and crowd mics, and a short delay to the lead vocal. I only used the crowd mics at the beginning and the ending of the track as they did not add any useful crowd participation, only a muddy echo of the PA during the song which reduced the clarity. I muted the tracks during the song and faded them back in when the crowd joins in. This is also an example of how you can edit introductions with cross fading to a song to make it flow and to add interest. 

​

Unfortunately, this song sounds drastically different to the recorded studio version as the acoustic guitar had its gain turned up way too much during the song and has given it a very distorted heavy electric guitar tonality. All I can do in the mix is work with it and emphasise that tone with EQ and reverb. Compression was not needed in this instance as the signal was already too squashed. I chose to use a gate on the floor tom as I had it set to a fast attack and slow hold and release as whenever the floor tom was played it was played continuously for a duration, so it stayed completely open during those sections and closed gently after these sections to mute the channel in between. I would not recommend this technique where one off hits occur on the floor tom in a song, like drum fills, as the opening and closing of a gate at that speed and for that duration can be heard and jarring to the energy and flow of a song. 

Day We Ran 2

For Day We Ran’s second show I was able to mix a full multi-track with crowd mics and have that separation and clarity to mix. These are some key differences to this recording compared the first one for the same song. 

 

Firstly, as this session was significantly larger than the first few, bussing together elements such as all the drum mics, guitars, vocals and crowd made automation and controlling volume levels much easier and more uniform. Increasing the volume of vocalists talking to the crowd and pulling back on reverb and delay effects is key when including these in sections before or after songs so they are clear, brought to the forefront, and grab the listeners attention on social media content, like I have done in their track Carolina. I have utilised muting channels such as the backing vocals when they are not actively singing, as I found the bleed picked up in those microphones were not beneficial to the mix and the processing I have used for the vocals reflected badly. Proximity effect was also strongly evident in the lead vocal here, so a high pass filter was used to start rolling off low end from 200Hz. All this additional bass in the vocal would cause the kick and bass to lose presence in the mix and reduce the clarity of the vocal. Elements such as the crowd mics, lead vocal, guitar, the grouped basses, and hi-hat were individually volume automated to maintain equal energy and volume level throughout, emphasise the crowd when they’re bringing the energy or take away some hi-hat when way too much crash cymbal bleed was unpleasant. I have once again emulated a bigger space on the multi-track with a room reverb on the vocals, guitars and crowd and short delay on the vocals. With a relatively clean multi-track for a live recording, I experimented with creative reverb and delay automation in the second verse to re-create the effect of the lead vocal echoing away in the original studio version and exaggerate the attempt of the vocalist to do this. I also felt a long fade out as the crowd was continually cheering and clapping was reflective of the nature that it was the last song of the set. All elements have been simply and subtly panned to reflect where the band members were standing on stage and create a sonic landscape that is expected from a listener. 

The Faith Estate

This recording has 0 separation between any tracks I had to work with and there was no ability to treat each element individually. The approach to this mix is to emphasise the capture of the FOH mix in the space rather than perfect the sonic balance of each element. I also believe that working with predominantly room recordings, having the FOH mix with all the vocal processing printed on to it adds to and works well with the room recordings. This is rather than having to replicate the vocal sound into the dry FOH signal and blend it with the other channels. I chose not to automate for this song either as all the tracks were loud and contained so much bleed, so the use of automation and excessive processing would have lost the energy of the audio. To improve the clarity strategic compression to control transients and attack and frequency balance channels to blend and mix made the difference. 

 

I found the relatively close room mic’s for this gig reduced the amount of extra noise and conversation that was picked up due to the loudness of the music in the space. Working with the engineer who knew the room, those recorders were placed where a lot of bass information was picked up from one and more high-end from the other. To emphasise this, I used a low pass filter on the track with more bass and a high pass filter on the one with more high-end information for extra separation and control of the frequency range. In saying this, I attenuated about 7.5dB of high mid information around the 6kHz range as the cymbals are very prominent in all audio channels and had to be tamed. The FOH LR stereo mix was mostly vocals, so I treated the processing like a vocal track with a UA 1176 compressor, a de-esser and an EQ, boosting about 2dB around 4kHz, and cutting about 1.5dB around 600Hz. I didn’t roll off low end like you would with regular studio vocal recordings as this signal contained extra kick and bass DI, however I made the compressor attack speed fast with an 8:1 ratio to tame the transients of the kick punching through these channels. 

​

Letters To Lions

The audio I captured from this show was a LR mix from the FOH desk and a stereo room recording from just above the desk. I only had 4 tracks to work with after I separated these stereo channels into mono ones to mix individually. 

 

Compared to the one stereo recording from Day We Ran, I used a lot of automation as this performance was very dynamic and energetic in Easy Honey. To consistently maintain that energy, automation brought it all to a fairy consistent level, blending the audio from the 2 different sources. The FOH mix was an independent mix I created with the live sound engineer to include the LR mix but with more kick, snare and bass D.I. I added a parallel compression send to these 2 tracks to add more saturation, punch and body to the mix of the close-mic’d elements. I also set up a reverb send to feed the room mics through. Separating the room mics into 2 mono channels was essential as the recorder was placed on its side to be secured to the wall, so one microphone pointed to the roof capturing the roof reflections and one pointed directly down to the crowd. These both had to be processed independently, as the tonality of them were quite different, with a lot more high-mid-range coming from the side pointing to the roof, while the one pointing down sounds much thinner and airier. I would advise not to record a room recording with this set up and find something where the stereo microphones are facing more directly to the stage. I still panned them left and right to maintain a stereo image, ensuring the tracks are still phase. With this mix, the snare drum is poking through a bit too much from the LR mix and the band mentioned how the bass player’s vocals were a bit too loud for their liking. Sadly, this cannot be adjusted at this point of mixing with the audio we have to work with, but this mix could have been different if a sound check recording had happened and had created a mix for myself before the show had started. 

Fat Salad

For the mixing of this set, I used LUNA, a DAW I am less familiar with but wanted to evaluate as a free alternative to DAWs such as Pro Tools or Logic. 

​

LUNA only has some stock hardware emulation UA plugins. Mixing in LUNA is not ideal for people used to mixing or using DAWs as it emulates hardware mixing with channel strips, signal flow and inserts, for example there is no parametric EQ. I recommend LUNA for a more sophisticated mixing process, however if you are new to mixing and DAWs, GarageBand is the most basic and straightforward free DAW to use to create a basic mix (I will also mention I had issues with LUNA crashing when trying to close or open different projects which as really annoying).

​

For this recording, no sound check was tested first and each track that was recorded into LUNA doesn’t quite match the FOH desk set up, indicating the multi-track recording patching from the FOH  desk was not correct. Additionally, in the tracks we do have to work with, there is so much bleed from everything on stage because of to the tiny stage for the 4 band members and loud guitar amps and drums. In this particular case, the band was using monitor wedges, that most artist playing these size indie-rock gigs do, however they were extremely loud and are feedbacking in the recording from the vocal mics and causing this phasing effect that you can hear when the band talks to the crowd during this song. This is one thing to be mindful of and can be checked during soundcheck. Unfortunately, the attempt to make a multi-track recording was unsuccessful as there is almost no separation between microphones to process each element individually. Proximity effect is prevalent in most tracks creating a lot of muddy low end without clarity or punch. I have only done some basic mixing with EQ, compression and reverb send for the room. The loudness of the performance is well reflected; however, I believe this mix and recorded audio was not ideal for bringing out clarity from the performance while keeping this energy. 

Conclusion

In summary, there is never one mixing formula, template, guide or rules. Your mix decisions will have to be subjective to the audio you have been able to capture and reflect the sound that you want your audience to know you by. In saying that, here are some key points I believe translate well in all kinds of indie-rock gig live recordings to bring clarity and maintain the energy you have brought in a performance, to the recording:

  1. Use a DAW you are most comfortable and confident to use.

  2. Don’t waste your time overdubbing, over sampling or stem splitting, choose sections of your performance that feel real and energetic and work with the audio you have, not against it.

  3. Compression – utilise many types of compression all over the mix to bring things forward, emphasise transients, squash transients, give sounds more body. 

  4. EQ - create more sonic separation of elements and bring forward fundamental frequencies and harmonics, roll off bass frequencies caused by proximity effect, reduce muddiness in the mid-frequency range between guitars and vocals. 

  5. Bring the space to life with a realistic room reverb.

  6. Glue the mix together with master bus processing – a soft bus compressor, tape saturation top warm up the mix, subtle EQ moves that enhance the saturation and a limiter. Ensure this master bus meets loudness level requirements for the platforms you want to share the audio to. 

 

Live recordings are not meant to be perfect, they’re meant to be real, don’t fight it. I hope this research project has made recording your own gigs more accessible, and a brought a new way to improve your social media presence and grow your audience. 

© 2035 by Patrick Thomas. Powered and secured by Wix

bottom of page