COMPLETE GUIDE

Pro Tools for Audio Post-Production: The Full Workflow

Session setup at 48 kHz/24-bit, dialogue editing, ADR recording, foley, music mixing, bus structure, iZotope RX exchange, LUFS metering and stem export for video sync. Every setting, every routing, every export step.

KKirillFounder & Creative Director

Hire an Audio Engineer See Our Work

Sections

48 kHz / 24 bit

Session format

-23 LUFS

EBU R 128 target

Avid Pro Tools is the digital audio workstation this workflow is built around: session setup, dialogue editing, ADR, foley, music mixing, loudness standard compliance and stem export. The guide covers the process end to end — from session settings to delivering finished files.

Session setup for audio post-production

Sample Rate and Bit Depth

Create a new session and set Sample Rate = 48 kHz, Bit Depth = 24-bit. Sample rate: 48 kHz is the default audio setting in DaVinci Resolve, and it is typical for broadcast and cinema work. Higher rates (96 kHz and above) take twice the processing power and produce media twice the size, so a 96 kHz session is not needed in a typical post workflow. Stay on 48 kHz unless the delivery specification explicitly asks for something else. Bit depth: 24-bit is the working value for post-production; in Fairlight it is the default, and 16- or 32-bit files can be selected at any time depending on delivery requirements. 32-bit makes sense for stem bounces that will be processed further. Audio file format: BWF (.WAV) — Broadcast Wave Format. In Fairlight this format is chosen when exporting audio (BWAV), and the sample rate is taken from the project. The format matters when exchanging files with video editors (Premiere Pro, DaVinci Resolve) that need to sync audio to picture. Use BWF for post-production.

Session Parameters Checklist

Sample Rate: 48 kHz. Bit Depth: 24-bit. Audio format: BWF (.WAV). Timecode rate: matches the video edit (23,976 fps, 25 fps or 29,97 fps — follow the video editor's timeline). Session start time: follow the video edit timecode so sound and picture start in the same place. Track timestamps: Timestamped (absolute timecode position). Turn on Elastic Audio on dialogue tracks — time compression and stretching without a pitch shift.

Bus Structure and Routing

Post-production sessions use a structured bus hierarchy. Set up the buses before importing any media: Aux Input (buses): - DX BUS: all dialogue tracks are routed here. It receives sound from the boom mic, lavalier mics and static microphones. - FX BUS: all sound effects and foley are routed here. - MX BUS: all music tracks are routed here. - BG BUS: all backgrounds and ambience are routed here. - STEM MSTR: the master bus that receives DX, FX, MX and BG. This is the print bus for the final mix. Why route through aux inputs instead of straight to the master? Because each bus gets its own insert chain. You can apply compression, EQ and limiting on the dialogue bus without touching the music bus. Create the buses: Setup > I/O Setup > Bus tab. Name them DX, FX, MX, BG, STEM MSTR. Assign colours: DX = green, FX = blue, MX = orange, BG = purple, STEM MSTR = red.

Dialogue editing

Dialogue Edit Workflow

Dialogue editing is the foundation of audio post-production. The goal: clean, intelligible dialogue with a natural room tone, smooth transitions between takes and no audible splices. Import the guide track (the mixed sound from the video editor's timeline) — it is your sync reference. Lock that track (right-click > Lock) so it does not shift by accident. Import the production sound: boom mic, lavalier mic and any static mic tracks. They arrive either as poly WAV files (several channels in one file) or as separate mono WAV files. If a file is poly, split the channels onto separate tracks (Track > Split to Mono). Checking sync: play the production sound alongside the guide track — they must match. If sample rates differ, watch the resampling: in DaVinci Resolve imported audio files at a different rate are automatically resampled to the timeline rate so they play correctly. Edit passes: - Comping: pick the best take for every line. If the boom is better for one line and the lavalier for the next, crossfade between them (Setup > Preferences > Editing > Default Crossfade = Equal Power). - Noise removal: send problem areas to iZotope RX. Remove air-conditioning hum, traffic noise and room echo. - Filling gaps: every gap in the dialogue should be filled with room tone. Record a stretch of room tone on set (silence with the actors still) and drop it into every gap. Never leave digital silence — the human ear reads absolute silence as unnatural. - De-breathing: reduce (do not delete) loud breaths with clip gain. Deleted breaths leave audible holes. - De-plosive: remove plosives (P, B, T overloading the microphone). According to the Fairlight guide, dialogue starts with a high-pass filter that rolls off everything below 85–100 Hz; then work the midrange around 250 Hz — the meatiest part of the voice, which often sounds boxy and closed.

Dialogue Editing Tips

Use Strip Silence (Edit > Strip Silence) to find and remove long silent stretches between lines quickly. Set the threshold and the minimum strip duration for your material: a threshold that is too high eats word endings, a minimum that is too short misses the pause.
Crossfade at every splice point. Even if two clips look perfectly aligned, a crossfade prevents clicks at the seam. Use a short crossfade for tight edits and a long one for smooth transitions.
Match room tone between takes. If the lavalier has more room reverberation than the boom in the crossfade, the transition is audible. Apply light reverb matching or clip EQ to balance the tonal difference.
Check dialogue in headphones after editing on monitors. Headphones reveal low-frequency rumble, mouth clicks and clothing rustle that nearfield monitors miss at low level.
Name every track and every clip. A session with names like DX_BOOM_TAKE3_V2 helps nobody. Use consistent naming: DX_BOOM, DX_LAV1, DX_LAV2, DX_PLANT.

ADR (automated dialogue replacement)

Recording and Editing ADR

ADR replaces production dialogue that is unusable because of noise, technical problems or the performance. The actor re-records the lines in a controlled environment while watching the picture. Setting up the session for ADR: create a mono audio track for recording. Route the video output to a display the actor can see. Route the guide track (the original production sound) to the actor's headphones at a comfortable level. Set input monitoring to Input Only on the recording track so the actor hears themselves in real time. Recording process: - Play the clip with the original dialogue — the actor listens. - Start recording: the actor says the line, matching the original in timing and emotion. - Record several takes of every line — more takes, more options at edit time. - Move on to the next line. Editing ADR: the recorded ADR must sync with the actor's lip movements on screen. Use one of the following approaches: Elastic Audio: turn on Elastic Audio on the ADR track. Use the Selector Tool to move the stretch and compression markers and stretch or compress the sound to match the lip sync. It suits small timing corrections. Revoice Pro (Synchro Arts): a tool for automatic ADR alignment. Send the original production dialogue and the ADR take to Revoice Pro — it aligns the ADR to the original timing. VocALign (Synchro Arts): a lighter option for aligning one audio clip to another. Pick the guide (the original dialogue) and the take (ADR), press Align: VocALign compresses and stretches the ADR to match the guide timing. Matching room tone: ADR recorded in a treated booth sounds drier than production dialogue recorded on set. Apply convolution reverb on the ADR track using an impulse response captured on location (if available) or a similar acoustic space. Match the length of the reverb tail and the level to the production dialogue.

Foley and sound effects

Recording and Editing Foley

Foley is the performance of everyday sound effects added in post-production to improve the sound of a film. Footsteps, clothing rustle, handling props, closing doors — all of it is performed by a foley artist while watching the picture. Setting up the foley recording: you need a foley stage or a quiet room with different floor surfaces (concrete, wood, carpet, gravel). A matched pair of condenser microphones in an ORTF or XY configuration, raised above the performance surface. Record to a stereo track in the session format: 48 kHz, 24-bit. Three categories of foley: - Moves (Moves): clothing and fabric sounds. The foley artist moves the fabric in sync with the actor's movements on screen. Record to a dedicated MOVES track. - Footsteps (Footsteps): the foley artist walks on the matching surface in sync with the actor's steps on screen. Record to dedicated FS tracks (separate tracks for left and right footwear if the mix requires it). - Specifics (Specifics): individual sound effects — opening a door, a glass on a table, rustling paper. Record to dedicated SPEC tracks. Editing foley: trim every foley hit to the exact frame. Adjust clip gain to match the perspective of the frame (closer = louder, wider = quieter). Add perspective EQ: pull down the top end for distant sounds and lift presence for close ones — according to the Fairlight guide, the 3–6 kHz band carries vocal presence and the 1–3 kHz band carries intelligibility. Fades on every hit: a short rise and fall to take out handling noise.

Sound Effects Libraries and Layering

Not every sound has to be performed by a foley artist. Library sound effects work for off-screen sounds, background ambiences and effects that are impractical to perform. Recommended libraries: - Pro Sound Effects (PSE) Core: a general collection of sound effects for video. - Boom Library: specialist libraries for gunfire, cars, weather and interface sounds. - Sound Ideas: the original broadcast SFX library. Library contents and prices change — check the manufacturer's current price list. Layered sound design: professional sound design stacks several sounds to build one effect. A gunshot is not one sound but four layers: the crack (high-frequency transient), the boom (low-frequency body), the tail (reverberation of the space) and the reaction (breaking glass, debris, the character's reaction). Each layer occupies its own frequency range and has its own dynamics envelope. Stacking gives a richer, more convincing result than a single library effect.

Mixing and loudness metering

Setting Mix Levels

Mixing sound for video is balance and intelligibility, not loudness. Dialogue has to sit clearly on top of music and effects. Level hierarchy (relative to dialogue): - Dialogue: the anchor level everything else is built around. - Music under dialogue: below the dialogue — far enough not to compete with speech. - Music on its own, without dialogue: at full level, in line with loudness standards. - Sound effects: according to the action on screen — hard effects can be louder than dialogue. - Backgrounds and ambience: below the dialogue, at the edge of audibility. Mix workflow: start with the dialogue. Set the dialogue bus fader so peaks keep headroom for effects and music. Add the backgrounds and drop them under the dialogue. Add the music and set it so it does not compete with speech in the same range. Add the effects last and set them to the action on screen. EQ for clarity: according to the Fairlight guide, the 1–3 kHz band carries intelligibility, 3–6 kHz carries presence and 5–8 kHz carries sibilants. Lifting 1–3 dB around 3–5 kHz makes a voice brighter and more intelligible. Music and effects that compete with speech in that range should be ducked automatically under the lines — that is the dialogue-ducking trick.

LUFS Loudness Standards

LUFS loudness standards: every delivery platform has its own target level, and you have to check the specific specification before you deliver. - YouTube: -14 LUFS integrated (the YouTube target level; in Resolve this is set as the meter target level). - Netflix: -27 LKFS with dialogue gating; for Atmos they recommend true-peak limiters no higher than -2,3 dBTP, and loudness is checked via a 5.1 rerender. - Spotify: -14 dB LUFS per the ITU-R BS.1770 standard. Normalization is applied on playback; on upload the track loudness is not processed. Quiet masters are lifted with headroom in mind: a track at -20 dB LUFS with a maximum true peak of -5 dB FS is lifted only to -16 dB LUFS. - Apple Podcasts: about -16 dB LKFS with a tolerance of ±1 dB; the true peak must not exceed -1 dB FS. - Broadcast in Europe (EBU R 128): -23 LUFS integrated, tolerance ±1 LU; the true peak must not exceed -1 dBTP. - Broadcast in North America (ATSC A/85): -24 LKFS integrated. Metering: Resolve has the loudness meters built in — M measures the equivalent level in a 400 ms window updated every 100 ms, Short measures the average level over 30 seconds, and Integrated reports the value for the part of the mix played through. On other workstations the LUFS meter is installed as a separate plug-in on the master bus. How to measure: play the whole mix from start to finish with the LUFS meter active — the integrated reading must land on the target. For orientation: according to the Fairlight guide, the true peak in a typical dialogue film sits between -6 dBTP and -3 dBTP, and the target maximum for normalization in Fairlight is -2,0 dBTP with a target loudness of -23 LKFS. If the mix is too loud: pull the master fader down and rebalance. If the mix is too quiet: do not raise the master fader — that raises the noise. Apply a limiter on the master bus instead, to control the peaks and lift the average level.

iZotope RX integration

Pro Tools to iZotope RX Roundtrip

iZotope RX integrates with Pro Tools in two ways: the RX Connect plug-in, and standalone RX with manual file exchange. RX Connect (preferred): insert RX Connect on the audio track you need to clean. Click the RX Connect plug-in to open the RX editor with the selected audio. Run de-noising, De-hum, De-reverb or any other RX module. Press Send Back in RX — the processed audio returns to Pro Tools and replaces the original. This workflow is non-destructive: RX Connect creates a new clip, and you can undo if the result is not what you wanted. Standalone RX: export the audio clip from Pro Tools (right-click > Export Clips as Files). Open the exported file in standalone RX, process it with any RX module, export the cleaned file, import it back into Pro Tools and replace the original clip on the timeline. Main RX modules for post-production: - Voice De-noise: real-time de-noising. Set Learn to capture a noise profile from a silent section, then adjust Reduction for the material. Heavy reduction produces audible artefacts — it is better to run the module twice than to push the reduction to the limit. - De-hum: removes electrical hum. Mains hum has a frequency of 60 Hz in America and 50 Hz in other parts of the world; De-Hummer has presets for 50 Hz and 60 Hz, and the default fundamental is 50 Hz. For a precise removal that does not touch neighbouring frequencies, keep the filter band narrow. - De-reverb: reduces room echo. Raise Reduction gradually — higher values give watery artefacts. - Mouth De-click: removes smacks and lip clicks. More sensitivity catches more clicks, but raises the risk of removing legitimate transient consonants. - Spectral Repair: manual repair of frequencies for isolated noise (a door slam during a line, a phone ringing). Select the noise on the spectrogram and press Replace — RX interpolates the surrounding sound to fill the gap.

Stem export for video sync

Bouncing Stems

Stems are separate mix buses exported as separate audio files. Video editors, colourists and broadcast engineers need stems for final corrections without a full remix. The stems required for standard delivery: - DX STEM: the dialogue mix (boom + lavaliers + ADR, processed) - FX STEM: the sound effects and foley mix - MX STEM: the music mix - BG STEM: backgrounds and ambience - M&E STEM: music and effects without dialogue — required for international distribution - FULL MIX: all stems together (the full mix) Bouncing the stems in Pro Tools: - Solo the bus you need (DX BUS) and run File > Bounce Mix. Format: BWF (.WAV). Sample Rate: 48 kHz. Bit Depth: 24-bit. Channels: mono for DX, stereo for FX/MX/BG/M&E, stereo for the Full Mix. File type: Interleaved (a single file for stereo) — that is how Fairlight writes a multichannel file. Name: ProjectName_DX_STEM_48KHZ24BIT_01.wav. - Repeat for every stem. - Bounce the Full Mix last, with all buses and no solos. Timecode: every bounce must start on the same timecode as the video — otherwise the stems will not sync when imported into the video editor. Head and tail: leave a little silence before the start and after the end of the programme so the video editor has room for crossfades at the timeline edges.

Standard Stem Naming Convention

ProjectName_DX_STEM_48KHZ24BIT_01.wav (Dialogue, 48 kHz, 24-bit, version 1)
ProjectName_FX_STEM_48KHZ24BIT_01.wav (Effects)
ProjectName_MX_STEM_48KHZ24BIT_01.wav (Music)
ProjectName_BG_STEM_48KHZ24BIT_01.wav (Backgrounds)
ProjectName_ME_STEM_48KHZ24BIT_01.wav (Music and effects, no dialogue)
ProjectName_FULLMIX_48KHZ24BIT_01.wav (Full mix)
Attach a text file (ProjectName_AudioDelivery_Notes.txt) with: sample rate, bit depth, start timecode, duration, LUFS measurement, known issues.

Equipment and monitoring

Monitoring for Post-Production

Accurate monitoring is what lets you make mixing decisions that will sound right on any playback system. You need two monitoring paths: nearfield monitors and headphones. Nearfield monitors: put them on stands and build an equilateral triangle with the listening position, tweeters at ear level. Calibrate the monitoring level with pink noise and a sound level meter: without calibration, level comparisons between sessions do not carry over. Headphones: use closed-back headphones for dialogue editing and noise assessment. Check the mix in headphones after every major session: headphones reveal problems that monitors miss at the edges of the frequency range. Subwoofer: add one if your content has significant low-frequency content (explosions, bass-heavy music, transport sounds), and calibrate it relative to the main monitors. Acoustic treatment: even the best monitors sound wrong in an untreated room. Bass traps in the corners come first, then absorbers at the first reflection points on the side walls and a diffuser panel on the back wall. Without treatment, mix decisions are made against the resonances of the room rather than against the sound itself.

Pro Tools keyboard shortcuts for audio post

Essential Post-Production Shortcuts (Mac / Windows)

Action	Mac	Windows
New session	Cmd+N	Ctrl+N
Import audio	Cmd+Shift+I	Ctrl+Shift+I
Bounce mix	Opt+Cmd+B	Alt+Ctrl+B
Separate clip at selection	Cmd+E	Ctrl+E
Trim to selection	Cmd+T	Ctrl+T
Fade (default)	F	F
Crossfade	Cmd+F	Ctrl+F
Zoom to selection	Opt+Cmd+[	Alt+Ctrl+[
Toggle solo	S (with key focus)	S (with key focus)
Toggle mute	M (with key focus)	M (with key focus)
Cycle edit modes (Shuffle/Spot/Slip/Grid)	number-row keys	number-row keys

Honest limitations of Pro Tools for audio post

What Pro Tools Does Not Do Well

Exchange with external analysis plug-ins. iZotope RX, Melodyne and SpectraLayers work with Pro Tools through file exchange or through RX Connect, not through direct random access to audio in real time. For quick fixes, offline rendering is more practical; for small ones, manual clip editing.
Video playback quality. The built-in video engine plays reference video alongside the sound, but heavy files have to be transcoded: Apple ProRes 422 HQ is a 10-bit codec with a target data rate of about 220 Mbit/s at 1920×1080 and 29,97 fps, and that much reference video is harder to play back than compressed H.264.
Spatial audio. The Atmos mixing workflow in Pro Tools is cumbersome compared with the integrated panner in Nuendo — for serious Atmos work, look at Nuendo or DaVinci Resolve Fairlight.
Collaboration. Pro Tools projects are not built for real-time multi-user editing: cloud collaboration exists, but for professional deadlines it is treated as a risk rather than the main scenario.
Cost. A Pro Tools Studio subscription costs money, and the price depends on the terms and the region — Avid publishes the current prices and upgrade terms. Alternatives with a one-off licence are cheaper to enter, but they give up some post-production features.

iZotope RX Dialogue Cleanup Guide DaVinci Resolve Color Grading Guide Resolve vs Premiere Pro Sound Design Cost Breakdown After Effects VFX Workflows Guide

Pro Tools Audio Post-Production FAQ

What sample rate should I use for audio post-production?

48 kHz at 24-bit. That is the typical rate and depth for broadcast and cinema work: in DaVinci Resolve audio is processed at 48 kHz by default, and in Fairlight the default bit depth is 24-bit with 16- or 32-bit files available. You only need 96 kHz where the delivery specification explicitly asks for it: it doubles the processing power and the media size. Use BWF (.WAV) for exchange with video editors.

How do I integrate iZotope RX with Pro Tools?

Use RX Connect plugin on the track you want to process. Click the plugin to send audio to the RX editor, process it with De-noise, De-hum, De-reverb, or other modules, then click Send Back to return the processed audio to Pro Tools. This roundtrip workflow is non-destructive and preserves the original clip.

What LUFS level should my final mix target?

It depends on the platform. YouTube: -14 LUFS. Netflix: -27 LKFS (dialogue-gated normalisation). Apple Podcasts: about -16 dB LKFS with a tolerance of ±1 dB. Broadcast in Europe (EBU R 128): -23 LUFS, true peak no higher than -1 dBTP. Broadcast in North America (ATSC A/85): -24 LKFS. Always check the specific delivery specification of the platform before mastering, and measure the integrated loudness over the whole programme.

What audio stems do I need to deliver for video production?

At least six stems: DX (dialogue), FX (sound effects), MX (music), BG (backgrounds), M&E (music and effects without dialogue, for international distribution) and Full Mix (the full mix). All at 48 kHz/24-bit BWF, with the same start timecode as the video, and with short tails at the edges.

Pro Tools or DaVinci Resolve Fairlight for audio post?

Pro Tools, if you work in a studio that uses it, or if you need to exchange sessions with other studios. Fairlight, if you are already in the Resolve ecosystem and want to avoid the cost of a Pro Tools subscription: Fairlight has built-in loudness meters (M, Short, Integrated), BWAV export and normalization targets, and it handles many post-production tasks. The choice is first of all a question of which environment your team already works in.

How do I remove 50Hz electrical hum from location audio?

Use the iZotope RX De-hum module. Mains hum has a frequency of 60 Hz in America and 50 Hz in other parts of the world — depending on where you shot, set the fundamental to 50 Hz or 60 Hz. De-Hummer has ready presets for both frequencies. Keep the filter band narrow so neighbouring frequencies are not affected, and start with light reduction: too strong a De-hum makes dialogue thin and phasey.

Need Professional Audio Post-Production?

Our Pro Tools engineers handle dialogue editing, ADR, foley, sound design, mixing and stem delivery to loudness standards for music videos, advertising, short films and long-form content. A free audio consultation.

Get a Free Audio Consultation