Boris FX CrumplePop Pro 2025.5 [WiN-MAC]

• Product: CrumplePop Pro
• Publisher: Boris FX
• Version: 2025.5.3
• Format: Standalone, VST3, AAX, AU
• Requirements: Windows 10 or later, macOS 11.0 or Later
• Source: https://borisfx.com/products/crumplepop/

Boris FX CrumplePop is an AI-powered audio restoration toolkit designed for content creators, combining specialized neural network models trained to identify and remove specific audio problems: background noise (AudioDenoise), room echo/reverb (EchoRemover), wind noise (WindRemover), traffic noise (Traffic Remover), electrical hum, plosives (PopRemover), rustling (RustleRemover), and clipped distortion (ClipRemover), plus breakthrough Voice Enhance for instant vocal clarity. It offers real-time processing via plugins (VST3, AAX, AU) across Adobe Premiere Pro, Audition, Avid Media Composer, Pro Tools, DaVinci Resolve, Final Cut Pro, Logic Pro, and GarageBand, plus standalone SoundApp application with exclusive AI models (Outdoor-Wind, Outdoor-Traffic) and batch processing. All processing occurs on-device with no cloud upload; one license grants access to all supported hosts. Designed for video editors, podcasters, and content creators pursuing studio-quality audio without audio engineering expertise, it addresses the need for fast, effective audio restoration enabling professional results in minutes rather than hours.

Key Takeaway

Boris FX CrumplePop transforms “audio cleanup” from a time-consuming technical task to a one-slider operation. For content creators drowning in background noise, echo, wind noise, and plosive problems, it’s transformative—what took hours of manual EQ, gating, and compression takes minutes. For audio professionals, it’s supplementary; for everyone else, it’s essential.

Problem-Specific Models, Not Knob-Tweaking

Traditional audio cleanup requires mixing-console expertise: load an EQ, identify problem frequencies, apply compression, add gating, adjust release times. CrumplePop inverts this entirely.

Instead of parameters, CrumplePop offers dedicated AI models. Each model (Echo Remover, Wind Remover, Traffic Remover) is a neural network trained on thousands of audio samples to identify and remove a specific problem. You don’t set threshold or ratio—you select the model and adjust a single Strength slider (0-100%).

The AI intelligently differentiates: “This is background noise, this is the voice, remove only the noise. This is room echo, this is the spoken word, remove only the echo.” This is fundamentally different from traditional plugin approaches and explains why users report “what used to take hours takes minutes.”

Voice Enhance: The Breakthrough New Tool

Voice Enhance is CrumplePop’s latest innovation—an AI model that enhances vocal clarity regardless of recording environment. Three variants (Normal, Bright, Warm) enable tonal adjustment.

The designers recommend using Voice Enhance first in any restoration workflow: enhance the vocal, then apply specific cleanup (echo removal, noise removal, etc.). The practical result: clearer, more professional-sounding dialogue instantly.

This is not traditional compression or EQ. It’s AI understanding what a human voice should sound like and enhancing it intelligently.

AudioDenoise: Removing the Hum, Hiss, Whirr

Vacuum cleaners, AC units, electrical hum, background noise—AudioDenoise removes persistent background sounds using AI. Unlike noise gates (which create pumping artifacts) or EQ (which sacrifices fidelity), the AI understands the problem: “That’s noise, this is signal. Remove the noise.”

Frequency-based controls (Low/Mid/High dials) enable fine-tuning: if the denoise is affecting the vocal tone, reduce the specific frequency band.

The result: clean audio without robotic artifacts.

EchoRemover: Untreated Rooms Made Professional

One of the most common creator problems: recording in an untreated room with echo/reverb. EchoRemover specifically targets this.

The AI automatically detects room echo and removes it. Controls (Strength, Dryness, Body, Tone) enable customization. ToolFarm’s review showed dramatic before/after transformation: a room-echo-heavy recording became clear and direct.

For video editors shooting interviews in unpredictable locations, this is a lifesaver.

WindRemover & Traffic Remover: Outdoor Solutions

The 2025 update adds two new AI models addressing outdoor recording challenges. WindRemover removes wind noise, including extreme conditions (rain, gusts). Traffic Remover reduces urban noise (cars, trucks, buses) while preserving vocal presence.

These address real creator pain: outdoor shoots always have compromised audio. Now creators can record anywhere without environmental limitations.

ClipRemover: AI Reconstruction of Distorted Audio

When someone shouts into a microphone and causes clipping (audio distortion), ClipRemover uses AI to reconstruct what was lost. This isn’t traditional bit-depth recovery—it’s AI understanding how the waveform should have looked and restoring it.

This is genuinely innovative: previously, clipped audio was considered unrecoverable. CrumplePop changes that.

PopRemover & RustleRemover: Detail-Level Cleanup

PopRemover targets plosives (P, B, T, D sounds) that pop into microphones. RustleRemover removes rustling and clothing noise from lavalier microphones.

These are details, but details matter. A “perfect” interview with one prominent plosive or constant rustling sounds unprofessional. CrumplePop removes these details.

Multi-Host Support: One License, Every Application

One CrumplePop license works in Adobe Premiere Pro, Audition, Avid Media Composer, Pro Tools, DaVinci Resolve, Final Cut Pro, Logic Pro, and GarageBand.

This is significant value: instead of buying separate audio tools for each application, one purchase covers everything. For editors switching between software, this is enormous convenience.

SoundApp: Standalone Processing

The 2025.5 overhaul adds a redesigned standalone application, SoundApp, enabling audio processing outside your NLE.

Import a clip, select AI model, adjust strength, export. That’s it. No plugin, no host, no project context required. Ideal for podcasters, pure audio creators, batch processing.

SoundApp also includes two exclusive AI models (Outdoor-Wind, Outdoor-Traffic) not available as plugins. The Model Manager enables downloading new AI models directly in the app—future-proof for new model releases.

On-Device Processing: Privacy & Security

All audio processing occurs on your computer. No audio uploads to cloud servers. No data collection. No external processing.

For creators working with sensitive material (confidential interviews, private lectures), this is essential peace of mind.

Real-Time Preview: Instant A/B Comparison

Toggle the effect on/off in real-time, hearing the difference immediately. This isn’t theoretical—users can instantly judge if the removal strength is appropriate.

This feedback loop is crucial: audio cleanup is subjective. Some prefer aggressive noise removal (risking artifacts); others prefer subtle preservation (risking remaining noise). Real-time preview enables informed decisions.

Not Magic, Just Very Good

CrumplePop is not magic. If audio is “90% noise, 10% signal,” removal will struggle. If echo is louder than the voice, removing it completely risks losing the voice too.

The principle: CrumplePop works best on typical creator audio problems (moderately noisy rooms, light wind, minor plosives). It’s not designed to salvage completely unusable audio.

Additionally, over-application of strength can add subtle robotic artifacts—the cure becomes almost worse than the disease. Recommended approach: use 70-80% of maximum strength, not 100%.

Pros	Cons
AI-powered simplicity (one-slider operation); no audio engineering knowledge required.	Not magic; works best on moderate problems, not extreme degradation.
Effective results on typical creator audio problems (echo, noise, wind, plosives).	Over-application adds subtle robotic artifacts; requires restraint.
Multi-host support (one license, all major NLEs + standalone).	Subscription model; perpetual licenses less featured.
Real-time preview enables instant A/B comparison.	Niche for audio professionals (supplementary, not replacement).
On-device processing (no cloud, privacy/security guaranteed).	New AI models (Wind, Traffic) less battle-tested than established tools.
Standalone SoundApp for audio-only creators.	No traditional parameter control (EQ, compression ratios) for advanced users.
Continuous AI model updates (Model Manager).	Pop Remover and ClipRemover work better on short selections.
Fast workflow (minutes, not hours).	Brand new Voice Enhance (2025) lacks extensive user feedback.
Professional results without hiring audio engineer.	Overkill for professionals; underkill for heavily damaged audio.

FAQs

Final Verdict

Boris FX CrumplePop is transformative for its target audience: video editors, podcasters, content creators pursuing studio-quality audio without audio engineering expertise. The AI approach fundamentally simplifies audio restoration. What took technical knowledge and hours takes minutes and a single slider.

For professionals, it’s a supplementary tool—useful for initial cleanup, but not replacing established professional audio tools like iZotope RX or mixing console techniques.

The subscription model is reasonable at $180/year; the multi-host support provides significant value. The 2025.5 update (Voice Enhance, SoundApp redesign, new AI models) shows active development and commitment to the product.

The main risk: brand newness of Voice Enhance (2025) and Traffic Remover (2025) means less real-world testing. Established tools like EchoRemover and AudioDenoise are well-proven.

For anyone frustrated by noise, echo, wind, and audio cleanup complexity, try the free trial. The before/after will justify investigation.

Rating: 4.5 / 5

AI-powered audio restoration toolkit spanning nine specialized models (Voice Enhance, AudioDenoise, EchoRemover, WindRemover, Traffic Remover, ClipRemover, PopRemover, RustleRemover). Multi-host support, standalone SoundApp application, on-device processing, real-time preview. Effective for typical creator problems; not magic for heavily damaged audio. Subscription model; perpetual licenses available. For content creators, essential; for professionals, supplementary.