# PublishMode User Manual PublishMode is a voice-first production chain for sung vocals, speech, podcasting, voiceover, streaming, and dialogue. Its cleanup, pitch, de-essing, and AUTO tools are designed around the behaviour of a human voice, while its EQ, dynamics, tape, ambience, and output stages can also be useful on other sources. ## Quick Start 1. Insert PublishMode on a vocal or voice track. 2. Select `LIVE` while recording or monitoring through the plugin. 3. Select `POST` while editing, mixing, or rendering. 4. Play a clear, representative phrase and click `MASTER AUTO` for a conservative starting point. 5. Adjust the enabled sections by ear. 6. Use the section `AUTO GAIN` switches (Mic Pre, EQ, Dynamics, Tape) and Output `AUTO` where appropriate so louder is not mistaken for better. AUTO tools only make a decision when they hear suitable material. A tool may leave its settings unchanged when the evidence is weak; this is intentional. ## How PublishMode Thinks PublishMode follows a practical studio order: prepare the recording, remove unwanted sound, shape tone, control movement, add colour and space, then protect the output. Most processes are deliberately conservative at normal settings so several small improvements can work together without erasing the performance. `Tone`, `Cleanup`, and `Colour` divide EQ into three distinct jobs. Tone makes broad musical moves. Cleanup dynamically reduces mud and boxiness only when those areas build up. Colour adds guarded warmth, vowel richness, presence, and air without blindly boosting noise, plosives, or sibilance. Mic Pre and Tape are also separate jobs. Mic Pre adds preamp-style density and focus; with its `AUTO GAIN` on (the default) it continuously keeps its output close to its input loudness. Tape is a complete machine stage: it has subtle oxide, hysteresis, head, and gap character even when its user controls are at zero. Every gain correction inside the machine itself is calibrated and fixed, so `SAT`, `PUSH`, and `BUMP` never chase the vocal level or pump with phrases; Tape's `AUTO GAIN` adds an optional, deliberately slow loudness match around the whole section so those controls read as colour instead of volume. The Vocal Suite adds tuning, harmony, character shaping, delay, and reverb. Delay and Reverb behave like send effects: the dry lead remains at unity, while their `Mix` controls add a wet return. `DUCK` compresses only that wet return while the lead is present, like a professional vocal sidechain setup. It does not turn down the lead vocal. Delta modes are reality checks. They audition what a cleanup process removes. If Delta contains too much recognizable vocal, reduce or retune the process. Only one Delta audition can be active at a time. ## Signal Flow The main processing order is: 1. Input 2. Clean 3. Vocal Tuning 4. Mic Pre 5. EQ 6. De-Esser and Dynamics 7. Tape 8. Harmony 9. Character 10. Delay 11. Reverb 12. Output and Limiter The `PRE`/`POST` switch in De-Esser changes its position around Dynamics: `PRE` de-esses before Dynamics, while `POST` de-esses after it. Tape follows both. ## Installation PublishMode beta is supplied as VST3 and AU on macOS and VST3 on Windows. Close the DAW before replacing an existing build. Common plugin locations: - macOS VST3: `~/Library/Audio/Plug-Ins/VST3/` - macOS AU: `~/Library/Audio/Plug-Ins/Components/` - Windows VST3: `C:\Program Files\Common Files\VST3\` After copying the plugin, open the DAW and rescan its plugin database. If an older build keeps loading, close the DAW, check that there is only one PublishMode bundle in its search paths, replace that copy, and perform a full rescan. ### REAPER does not find PublishMode 1. Close REAPER. 2. Confirm that `PublishMode.vst3` is in a VST3 folder included in REAPER's VST paths. The standard macOS user folder is listed above. 3. Reopen REAPER and go to its VST plug-in settings. 4. Verify the VST path, then use the full cache-clear/re-scan option rather than a quick scan. 5. Search for `PublishMode` under VST3. On macOS, it may also appear under AU after an AU rescan. The macOS beta is a universal Intel/Apple Silicon build. If macOS blocks a trusted beta package, follow the instructions supplied with that exact package. Do not disable Gatekeeper globally or remove security attributes from unrelated files. ## System Requirements and Performance Minimum: - macOS 10.13 or later on Intel or Apple Silicon, or Windows 10/11 64-bit - A 64-bit VST3 host, or a 64-bit AU host on macOS - Dual-core 64-bit CPU from roughly 2015 or newer - 8 GB RAM - At least 768 pixels of vertical display space; the window scales down when needed Recommended: - macOS 12 or later on Apple Silicon, or a modern Windows quad-core CPU - 16 GB RAM - A 1920×1080 or larger display - A stable low-latency audio driver when tracking through `LIVE` Measured reference loads at 48 kHz on a 2021 Apple Silicon Mac are approximately: - Full `LIVE` chain at a 64-sample buffer: about 10% of one audio core. - Full `POST` chain, including cleanup, tuning, harmony, character, and 2× oversampling: about 20% of one audio core, plus pitch work on background cores. - Typical POST dialogue chain with Clean, EQ, De-Esser, Dynamics, Tape, and Limiter: about 12% of one audio core. These are reference measurements, not guarantees for every host or session. Sample rate, buffer size, active pitch engines, and instance count all matter. PublishMode is tested at 44.1, 48, 88.2, and 96 kHz. Higher rates use proportionally more CPU. PublishMode offers `OFF` and `2X` oversampling. At 48 kHz, `2X` in POST is the normal final-quality choice. Changing oversampling re-prepares the processing chain and may briefly mute the insert; change it while stopped when possible. ## Main Window ### Header - `MONO` / `STEREO`: chooses mono or stereo input handling. - `LIVE` / `POST`: changes the latency and quality strategy for the entire plugin. - Quality display: shows `Low latency • OS OFF` in LIVE or the current POST quality and oversampling state. In POST, click the display to switch oversampling between `OFF` and `2X`. - Gear: opens factory and user presets, A/B storage, oversampling, meters, defaults, tooltips, and Cat Comments. - `FX`: opens the Vocal Suite. ### Scope The main scope displays level, spectrum, processing activity, and context-specific graphics. - `IN` / `OUT`: displays analyser data before or after the chain. - `HOLD`: freezes the analyser trace. - `MASTER AUTO`: starts the full voice setup pass. - `REVERT AUTO`: restores the state saved immediately before the last Master Auto pass. When EQ is selected, the scope is also an editor: - Double-click empty space to add a custom EQ band in Tone or Colour mode. - Drag a band node to change frequency and gain. - Scroll over a node to change Q. - Right-click a node to choose Bell, Low Shelf, High Shelf, Low Cut, or High Cut. - Double-click a custom node to remove it; double-click a built-in node to reset it. In EQ Cleanup mode, the large `MUD` and `BOX` handles can be grabbed directly and moved horizontally to place each cleanup centre. Their wide ranges let either handle move freely across the useful analyser area and cross the other handle. The handle positions are user settings: audio playback and changes to the Mud or Box reduction amount do not move them. Use `TUNE` to move both cleanup centres together. When Tape is selected, the scope shows the machine response and drive activity, including visual feedback for `SAT`, `PUSH`, and `BUMP`. ### Section Switches and AUTO State Every main section has its own on/off switch. Turning a section off also cancels that section's active AUTO listening state, so a stale `RUN AUTO` message does not remain over an inactive section. Disabled sections do not process audio. Tooltips explain controls under the pointer. Cat Comments add short context-sensitive suggestions and can be disabled from the gear menu. ## LIVE and POST Use `LIVE` while recording, monitoring, streaming, or broadcasting. Use `POST` while editing, mixing, analysing, and rendering. `LIVE` adds zero samples of reported processing latency. It forces oversampling off and uses zero-lookahead versions of Tuning, De-Esser, and Limiter. Clean is Gate-only in LIVE. Deroom, Denoise, Click, De-Breath, Harmony, and Character are POST-only and appear locked or veiled instead of silently pretending to work. `POST` enables the complete processing chain and reports its latency to the host for plug-in delay compensation. Harmony and Character align their wet processing to the lead in this mode. Always switch back to POST before a final mix or render. ## Master Auto Master Auto performs a staged setup rather than changing every control at once. It establishes input gain first, listens to the corrected signal for cleanup and tone decisions, then finishes with dynamics, colour, and output matching. To use it: 1. Select a normal, representative part of the performance. 2. Click `MASTER AUTO`. 3. Play clear voice for several seconds. 4. Let the pass finish, then stop playback. 5. Review every enabled section and adjust by ear. Depending on the material and available modes, Master Auto can set Input gain and filtering, Gate, Clean starting points, Mic Pre, all three EQ jobs, De-Esser, Dynamics, Tape, and Output. It does not force a De-Esser decision without real sibilance or invent a cleanup profile from silence. Click `REVERT AUTO` to restore the complete state from before the pass. ## Input Input prepares the signal before cleanup and tone processing. Controls: - `GAIN`: level into PublishMode. - `HPF`: high-pass cutoff for rumble, handling noise, and unnecessary low end. - `LPF`: low-pass cutoff for unwanted high-frequency noise. - `FLAT`: disables both filters. - `LOWCUT`: sets a useful vocal high-pass starting point and disables the low-pass filter. - `PHONE`: creates a deliberately narrow telephone-style band. - `AUTO`: sets conservative gain and voice filtering from the material it hears. Moving HPF or LPF enables that filter. Use `FLAT` when you want both filters completely out of the path. Make Input decisions before compression and limiting; an overdriven source cannot be repaired by turning down Output. ## Clean Clean removes unwanted sound before pitch, EQ, compression, and saturation can exaggerate it. Select the required tool from its menu. Each tool has its own AUTO state and controls. ### Gate - `AMOUNT`: how firmly the background is reduced between phrases. - `ATTACK`: how quickly the gate opens. - `RELEASE`: how naturally it closes after words. - `AUTO`: listens for voice and establishes a safe floor and timing. Gate is the only Clean tool available in LIVE. If word beginnings disappear, reduce Amount or lengthen the release behaviour. ### Deroom - `AMOUNT`: room-reflection reduction. - `PRESENCE`: balances cleanup against retained vocal presence. - `AUTO`: analyses a representative passage and builds a room map. ### Denoise - `AMOUNT`: steady-noise reduction. - `PRESENCE`: balances noise reduction against retained vocal detail. - `AUTO`: analyses representative noise and voice and builds a noise map. ### Click - `AMOUNT`: sensitivity and strength for short mouth-click and impulse repair. - `AUTO`: analyses a representative passage for click events. ### De-Breath - `AMOUNT`: breath attenuation. - `NATURAL`: keeps reductions smoother and less binary. - `ML`: enables the trained breath classifier. - `AUTO`: analyses the performance and builds a breath map. Deroom, Denoise, and De-Breath keep their processing controls locked until their required analysis reports ready. Their analysis is track-specific and is saved with the session, not copied into general-purpose user presets. In REAPER, save the project before running a media-based Clean analysis so PublishMode can inspect the current project audio. If a host requires live capture instead, start shortly before a representative passage and play it through. Click `DELTA` to hear only the signal Clean is removing. If recognizable words, sustained vowels, or much of the musical pitch appear in Delta, lower the relevant Amount or rerun analysis on better source material. ## Mic Pre Mic Pre adds focused preamp colour before EQ and dynamics. - `TYPE`: `TUBE`, `TRANSFORMER`, or `SOLID STATE`. - `FOCUS`: `LOW`, `FULL`, or `HIGH`. - `DRIVE`: harmonic density and saturation. - `MIX`: clean/coloured blend. - `AUTO`: chooses a restrained voice-dependent starting point. - `AUTO GAIN`: continuously matches the section's output near its input loudness. It is on by default. Tube is smoother and more even. Transformer adds weight and punch. Solid State is faster, more forward, and can become edgier at high Drive. With `AUTO GAIN` on, Mic Pre gives steady gain-matched results so you judge Drive and Mix by tone rather than an apparent gain increase; turn it off when you want the raw level change of the preamp stage. ## EQ EQ has three modes. The mode menu changes both the controls and the job being performed. ### Tone - `BODY`: broad low-frequency weight. - `CLARITY`: presence and intelligibility. - `AIR`: top-end openness. - `AUTO`: establishes a conservative broad balance. The built-in Tone nodes can also be positioned in the scope. Start with broad moves before adding custom PEQ bands. ### Cleanup - `MUD`: dynamic low-mid reduction amount. - `BOX`: dynamic boxy-mid reduction amount. - `Q`: width of the cleanup bands. - `TUNE`: shifts both cleanup centres together. - Scope `MUD` and `BOX` handles: place the two centres independently. - `AUTO`: listens live for reliable mud and box evidence, then sets restrained starting points. - `DELTA`: auditions only the signal removed by Cleanup. Cleanup is dynamic: reduction increases when the problem becomes prominent instead of cutting the same hole through the whole performance. AUTO is intentionally cautious. If it does not hear reliable buildup, it leaves the current settings alone. Changes to Amount affect reduction depth only; they do not reposition the scope handles. ### Colour - `WARM`: guarded low-mid warmth. - `HNY`: vowel richness and body. - `BITE`: forward presence with harshness protection. - `SILK`: dynamic air with hiss and sibilance protection. - `CHARACTER`: `CLEAN` for polish or `DIRTY` for a more driven colour path. - `AUTO`: sets subtle additive sweetening from the voice it hears. Colour works best after obvious noise and resonances are already controlled. Use small amounts; additive tone can accumulate quickly. ### Shared EQ Controls - `AUTO GAIN`: continuously matches EQ output near its input for fair tone decisions. It is on by default. - Custom PEQ: up to eight bands can be created in the scope, with Bell, Low Shelf, High Shelf, Low Cut, and High Cut shapes. Only Cleanup supports EQ Delta. If Delta contains too much vocal body, lower Mud or Box, widen the band, or move its handle to the actual problem frequency. ## De-Esser De-Esser reduces focused sibilance without darkening the whole vocal. Essentials: - `AMOUNT`: maximum sibilance reduction. - `FREQUENCY`: centre of the detection band. - `PRE` / `POST`: before or after Dynamics. - `DELTA`: auditions only the detected reduction. - `AUTO`: listens for real sibilant events and sets a conservative starting point. Open `ADV` for Width, Sensitivity, and Release. PRE is useful when compression would exaggerate esses. POST is useful when compression or saturation creates a final bright edge. If Delta contains vowels or general brightness, lower Amount or Sensitivity, narrow Width, or retune Frequency. AUTO requires phrases containing sounds such as `S`, `SH`, or `CH`. It can correctly make no change when the passage has no trustworthy sibilance. ## Dynamics Dynamics combines a fast Peak compressor with a slower Level compressor. Essentials: - `INPUT PUSH`: `LOW`, `STUDIO`, or `HOT` drive/voicing into the dynamics chain. - `PEAK`: fast peak-control amount. - `LEVEL`: slower phrase-level control. - `FLOW`: switches between Peak → Level and Level → Peak. - `PARALLEL`: blends the controlled and direct paths. - `AUTO GAIN`: restores the average gain reduction slowly while voice is present and holds through pauses. - `AUTO`: sets Peak, Level, Flow, Parallel, and enables Auto Gain from the detected voice. Open `ADV` for independent Peak Ratio, Attack, and Release plus Level Ratio, Attack, and Release. Peak ratios use the stepped choices `4:1`, `8:1`, `12:1`, and `20:1`. AUTO GAIN is not a fast loudness rider. It makes up measured average compression loss with a deliberately slow response, so it does not chase each word. Disable it when you want to hear or keep the raw gain reduction. For a clean vocal starting point, Peak → Level usually works well; Level → Peak is denser and more controlled. ## Tape Tape models a complete record/playback path rather than adding a single saturation curve. When Tape is enabled, its oxide, hysteresis, head-gap response, and magnetic knee create a subtle machine sound even with every user control set to zero. Therefore, Tape enabled with all controls at zero is expected **not** to null against bypass. ### Basic Controls - `SPEED`: `7.5 ips`, `15 ips`, or `30 ips`. - `SAT`: adds asymmetric oxide/hysteresis harmonics above the baseline machine colour. - `PUSH`: raises operating level into a fixed magnetic soft knee, up to `+18 dB`, with calibrated electrical compensation. - `TONE`: adjusts the reproduce tilt darker or brighter. - `AUTO`: chooses a restrained speed, Sat, Tone, and tiny transport movement from vocal level and crest factor. - `AUTO GAIN`: slowly matches the whole tape stage's output loudness to its input. It is on by default and is deliberately much slower than the other section matchers. `SAT` and `PUSH` are independent. Sat changes the harmonic layer; Push changes how firmly the signal works the magnetic knee. Push has no detector, release, or adaptive makeup stage inside the machine, so it will not ride phrase loudness. `AUTO GAIN` sits around the whole section instead: it averages roughly a second of programme and moves its correction slowly in both directions, giving steady gain-matched results without pumping. Turn it off to keep the machine's calibrated raw levels, for example when you want Push to genuinely raise the vocal. ### Advanced Controls Open `ADV` to replace the Basic knobs with: - `BIAS`: record calibration. The centre `CAL` position is neutral; under- and over-bias add controlled colour. - `BUMP`: broad speed-dependent low-frequency head weight. `0%` adds none, `100%` is a calibrated full setting, and `200%` is an intentionally obvious extension. - `WOW`: slow transport drift. - `FLUTTER`: faster transport movement. - `LINK`: copies Wow into Flutter when enabled and then moves them together. Tape AUTO does not overwrite Push, Bias, or Bump. The three speeds provide different balances: 30 ips is tighter and more extended, 15 ips is balanced, and 7.5 ips is softer with broader low-end bloom. If Tape becomes cloudy, try 30 ips, reduce Bump, or brighten Tone before assuming Sat is the only cause. The Tape tooltips, Cat Comments, and scope display describe this current machine model. At high Sat, Push, and Bump settings, leave Tape `AUTO GAIN` on to compare at matched level, and check Dynamics Auto Gain or Output AUTO if the whole vocal still appears to move in level. ## Vocal Suite Click `FX` to open Tuning and the four creative modules: Harmony, Character, Delay, and Reverb. The Vocal Suite master switch and the individual module switch must both be on for a creative effect to process audio. Editing a module can enable that module, but the master state still determines whether the suite is heard. Use `ESSENTIALS` for the main controls and `ADVANCED` for the complete control set. ### Suite Presets The `SUITE PRESET` menu changes Vocal Suite settings only; it does not overwrite the complete main-chain preset. - `Default`: neutral manual tuning, a quiet Double, delay off, and a useful plate starting point. - `Natural Space`: Natural tuning, a subtle Double, dotted-eighth digital delay, Hall reverb, and strong wet-return ducking. - `Analog Warm`: Modern tuning, a subtle Double, synced eighth-note Tape delay, a warmer Plate, saturation, and moderate wet-return ducking. Suite presets enable the modules they require. They do not add hidden Character processing. ### Tuning Tuning runs after Clean and before Mic Pre. - `STYLE`: `Natural`, `Modern`, `Hard Tune`, or `Manual`. - `MODE`: `Auto`, `Scale Only`, or `Chromatic`. - `KEY`: C through B, or `Auto`. - `SCALE`: Major, Minor, Harmonic Minor, Melodic Minor, modal, pentatonic, or Chromatic. - `FIND KEY`: listens to a clear pitched passage for up to about eight seconds and writes the detected key and scale. - `AMOUNT`: correction strength. - `ATTACK`: how quickly correction engages or moves to a new target. - `SMOOTH`: transition smoothness. - `NATURAL`: retained pitch movement and expression. - `FORMANT`: vocal-character preservation or shift. - `RANGE`: maximum correction distance in semitones. - `RELEASE`: how correction lets go between notes. Essentials shows Amount, Attack, Smooth, and Formant; Advanced adds Natural, Range, and Release. Natural uses gentler correction and more preserved movement, Modern is firmer, and Hard Tune is intentionally fast and audible. Manual leaves your knob values in control. `Auto` follows the selected or detected scale while allowing obvious passing notes. `Scale Only` is stricter. `Chromatic` targets the nearest semitone without key filtering. With Key set to Auto, play enough stable pitched material before relying on scale-aware harmony. LIVE uses the zero-latency Live Safe tuning engine. POST uses the full finishing engine. ### Harmony Harmony creates up to three aligned pitched voices. - Voice choices: `Off`, `Oct Down`, `5th Down`, `3rd Down`, `3rd Up`, `5th Up`, `Oct Up`, and `Double`. - `V1`, `V2`, `V3`: individual voice levels. - `IN KEY`: makes thirds and fifths diatonic to the selected or stable detected key. - `MIX`: total harmony return. - `WIDTH`: stereo spread. Double uses a dedicated short, modulated second-take path. Pitched voices share a stable POST alignment so mixed stacks remain together. Octave Down can be demanding on breathy, noisy, or unpitched source material; feed it a clean monophonic vocal and avoid forcing tuning across note transitions. Harmony is POST-only. Start with a quiet Double for width, or use a low third/fifth layer after confirming the song key. Built-in stack compensation prevents extra voices from winning only because they are louder. ### Character Character is a practical voice-shaping tool rather than a fixed Monster or Robot effect. Voice Type presets: - `Manual` - `Full Body` - `Low & Warm` - `Light & Airy` - `Gentle Grit` Controls: - `FORMANT`: deeper or brighter vocal identity. - `PITCH`: semitone shift. - `SUB`: lower supporting layer. - `DRIVE`: controlled soft grit. - `BODY`: leaner/lighter to fuller weight without hidden pitch or growl. - `MIX`: processed/direct blend. Character is POST-only. Use small Formant, Body, Drive, and Mix moves for lead-vocal polish; larger Pitch or Formant moves are deliberate special effects. Character adds no continuous noise floor and uses level awareness to keep transformations near their input loudness. ### Delay Delay is a send-style wet return. Its dry lead path remains at unity. Delay-only presets: - `Manual` - `Vocal Slap` - `Dotted 8th` - `Quarter Throw` - `Dark Echo` Controls: - `MODE`: `Digital`, `Analog`, `Tape`, or `Slapback`. - `SYNC`: locks timing to the DAW tempo. - `TIME`: manual delay time when Sync is off. - `DIVISION`: `1/4`, `1/4.`, `1/8`, `1/8.`, `1/8T`, or `1/16` when Sync is on. - `FEEDBACK`: number and decay of repeats. - `TONE`: repeat brightness. - `SAT`: repeat saturation. - `DUCK`: compresses only the wet delay return from the dry vocal sidechain. - `P-PONG`: stereo repeat spread. - `MIX`: wet-return level. Strong Duck settings can move the delay return roughly 16 dB behind active words, then release it into the gaps. The lead itself is not reduced. Use Feedback and Mix to control clutter before making extreme tone changes. ### Reverb Reverb is also a send-style wet return with a unity dry lead. Reverb-only presets: - `Manual` - `Tight Plate` - `Vocal Plate` - `Intimate Room` - `Wide Hall` Controls: - `MODE`: `Room`, `Hall`, `Plate`, or `Chamber`. - `SIZE`: scale and tail length. - `PREDELAY`: time between the dry vocal and the reverb onset. - `TONE`: tail brightness. - `SAT`: tail density and colour. - `DUCK`: compresses only the wet reverb return from the dry vocal sidechain. - `WIDTH`: stereo spread. - `MIX`: wet-return level. Strong Duck settings can move the reverb return roughly 14 dB behind active words, then let the tail return between phrases. Predelay provides additional separation without changing the dry lead gain. ## Output Output controls final delivery and peak safety. - `TRIM`: final output adjustment. - `PUSH`: deliberate pre-limiter drive from `-9 dB` to `+9 dB`. - `CEILING`: limiter ceiling. - `LIMITER`: enables final true-peak limiting. - `AUTO`: enables a safe limiter setup, measures a short pass, and matches processed output loudness to the plugin input without raising a quiet source toward an arbitrary target. - `ADV`: opens Limiter Release and Lookahead. Click a lit Output AUTO button again to clear its measured correction. In LIVE, the limiter uses a zero-lookahead path. In POST, it uses the selected lookahead and reports the latency to the host. A ceiling near `-1.0 dB` is a useful general streaming starting point, but delivery specifications vary. Push is a creative loudness/density decision; if the limiter becomes obvious, reduce Push or earlier Dynamics rather than increasing Release blindly. While the Output section is enabled, a final hard safety path protects against samples above 0 dBFS even if the main Limiter switch is off. ## Presets, A/B, and Settings The gear menu contains: - Oversampling `OFF` and `2X`. - Reset meters. - Show or hide tooltips. - Enable or disable Cat Comments. - Store the current state as A or B. - Swap to the other A/B state. - Load factory presets. - Load or delete user presets. - Save a user preset. - Reset Defaults. Factory presets reset the chain to a known base before applying their settings: - `Init`: default starting state. - `Clean Voice`: restrained cleanup and intelligibility. - `Warm Broadcast`: denser, warmer spoken-voice finish. - `Natural Singer`: transparent sung-vocal polish. - `Vocal Suite`: creative vocal-suite starting point. - `Broadcast Voiceover`: controlled, present voiceover chain. - `Modern Lead Vocal`: polished contemporary lead. - `Pop Hard Tune`: stylised fast tuning and pop finish. - `Intimate Podcast`: close, controlled spoken-word sound. User presets store reusable parameter settings. Track-specific Clean analysis maps are session data and are intentionally not carried into a preset for another recording. Use A/B to compare two complete states in the same instance; store both before swapping. ## Practical Workflows ### Record or Monitor a Vocal 1. Select `LIVE`. 2. Set Input Gain manually or run Input AUTO on normal singing/speech. 3. Use `LOWCUT` or a gentle manual HPF only if needed. 4. Use light Gate and Live Safe Tuning if the performer needs them. 5. Keep Harmony, Character, and analysis cleanup for POST. 6. Return to `POST` before mixing or rendering. ### Mix a Lead Vocal 1. Save the session and select `POST`. 2. Run Master Auto over a representative phrase. 3. Review Clean first and use Delta. 4. Shape Mic Pre and EQ at matched loudness. 5. Tune De-Esser using a phrase with several clear esses. 6. Balance Peak and Level in Dynamics. 7. Add Tape, then Vocal Suite effects only as needed. 8. Use Output AUTO for a fair final bypass comparison. ### Analyze Noisy Dialogue 1. Select `POST` and save your project or session first. 2. Choose one Clean tool at a time. 3. Run its AUTO over representative room, noise, breaths, or clicks. 4. Wait until map-based controls report ready. 5. Raise Amount conservatively while checking Delta. 6. Add Gate only for the remaining gaps, not as a substitute for every cleanup tool. ### Tune a Sung Vocal Naturally 1. Select Natural style. 2. Set the known Key and Scale, or run Find Key on a clean pitched passage. 3. Start with moderate Amount, slower Attack, high Smooth, and high Natural. 4. Reduce Range if you only want nearby notes corrected. 5. Check transitions, scoops, vibrato, and held notes before increasing Amount. ### Add Clear Vocal Space 1. Open `FX` and select `Natural Space`, or start from the Delay and Reverb section presets. 2. Set Delay and Reverb Mix for the gaps between words. 3. Raise each Duck control until the lead remains clearly in front. 4. Adjust Predelay and Delay division for the song tempo. 5. Check the wet tails during pauses as well as during full instrumentation. ## Troubleshooting ### AUTO keeps showing after a section is off Turning a section off should cancel its active listening state and clear its AUTO overlay. If a host automation lane immediately turns the section or AUTO parameter back on, disable that automation and try again. ### AUTO says it needs voice - Play clear, normal singing or speech rather than silence, breaths alone, or a full mix. - Start playback shortly before the representative phrase. - Use a passage containing the evidence the tool needs; De-Esser needs esses, for example. - Save a REAPER project before Clean's media-based analysis. ### Cleanup sounds unnatural - Reduce the relevant Amount. - Check Delta for recognizable vocal content. - Move the Cleanup MUD or BOX handle to the real buildup instead of increasing reduction at the wrong frequency. - Rerun map analysis on cleaner, more representative material. ### Cleanup MUD or BOX handles move unexpectedly Their positions should follow only direct dragging, `TUNE`, AUTO, preset loading, host automation, or state recall. Audio playback and Amount changes do not intentionally move them. Check for automated Cleanup Tune/Mud Offset/Box Offset parameters if a handle still moves on its own. ### De-Esser AUTO makes no change Run it over several clear `S`, `SH`, or `CH` events. A no-change result on a passage without sibilance is safer than a guessed, over-aggressive setting. ### Dynamics changes the vocal level Compression normally reduces gain. Enable Auto Gain to restore average gain reduction slowly, or disable it when you want the raw compressed level. Auto Gain does not lock every syllable to the same loudness. ### Tape does not null with every control at zero This is expected. Enabling Tape engages its baseline machine path even at zero user settings. `SAT`, `PUSH`, `TONE`, `BIAS`, `BUMP`, `WOW`, and `FLUTTER` alter that base sound; they are not the on/off switch for it. ### Tape loudness seems to fluctuate Tape Push uses a fixed magnetic knee and calibrated compensation, not a voice-following detector. Tape's `AUTO GAIN` is the only level rider in the section, and it is deliberately slow; toggle it to confirm whether it is the source of a slow drift. Otherwise check Dynamics Auto Gain, Output AUTO, Delay/Reverb Duck, host automation, and the level before PublishMode. Also compare the dry vocal rather than the changing combined dry-plus-ambience level. ### The vocal is distorted - Lower Input Gain or the level feeding PublishMode. - Reduce Mic Pre Drive, Tape Sat/Push, Character Drive, or Output Push. - Check the source and upstream plugins for clipping. - Keep the Limiter enabled as final protection, not as a repair for an overloaded chain. ### The vocal is too dull - Check Input LPF and De-Esser Amount first. - Brighten Tape Tone or try 30 ips. - Reduce excessive Tape Bump if the balance is cloudy. - Add a small amount of EQ Clarity, Air, or Colour Silk. ### The vocal is too bright - Use De-Esser for sibilance instead of cutting all high frequencies. - Reduce EQ Air, Clarity, Bite, or Silk. - Darken Delay or Reverb Tone if only the ambience is bright. ### Duck makes the vocal seem quieter Delay and Reverb Duck do not lower the dry lead. They remove wet-return level while words are present, so the total dry-plus-wet meter reading can fall even though the lead stays unchanged. Bypass the ambience modules or set their Mix to zero to compare the dry lead, and check downstream Output processing if the dry level itself changes. ### Harmony is late, choppy, or unstable - Use POST and allow the host's plug-in delay compensation to update. - Stop and restart playback after changing mode or sample rate. - Feed Harmony one clean monophonic vocal rather than a full mix or stacked recording. - Set the correct key for scale-aware thirds and fifths. - Reduce other pitch processing when testing difficult octave material. ### Effects are not audible - Enable the Vocal Suite master switch. - Enable the individual module. - Raise its Mix and confirm the full plugin is not host-bypassed. - Switch to POST for Harmony and Character. ### Latency is too high while recording - Select LIVE. - Confirm that the quality display says OS OFF. - Disable host lookahead or high-latency plugins elsewhere on the monitored path. - Use POST only after recording. ## Best Practices - Gain-stage before cleanup, compression, saturation, and limiting. - Run AUTO on representative material, then review its decisions. - Use one Delta mode at a time and turn it off after checking. - Prefer several small corrections to one extreme process. - Compare tone and dynamics at matched loudness. - Use send-style Duck to clear space around the lead instead of lowering the lead itself. - Save reusable parameter presets, but keep track-specific cleanup analysis with the session. - Render the final result in POST and listen through once after any sample-rate or oversampling change. PublishMode works best when it supports the performance. Start clean, make deliberate moves, and let the voice remain recognizable as itself.