Can Krisp be vibe coded?
Real-time AI noise cancellation and meeting voice enhancement
You can build post-processing and maybe route audio through open models, but real-time low-latency virtual-device noise cancellation is not a one-sitting web app.
Jump to the build brief ↓Checked Jul 2026
What you pay today, before any DIY hosting
medium editorial confidence
Buildability by layer
Screens, forms, and focused interactions
The repeatable job the product performs
Availability and legality of required data
Uptime, queues, support, and maintenance
Security, compliance, and user confidence
The achievable core
- Use an open noise-suppression model as a virtual microphone or post-process recordings before sending them to calls or files.
- Build a focused single-user workflow with real persistence, search, and export.
- A responsive interface with real empty, loading, success, and error states.
The parts a prompt cannot buy
- low-latency virtual microphone
- polished app switching
- model quality
- meeting integrations
- The last 20 percent is sync, migration fidelity, speed, and edge cases.
- Model quality and inference operations are part of the product.
Why people still pay
They pay because audio cleanup must be instant and invisible during real calls.
The last 20 percent is sync, migration fidelity, speed, and edge cases.
Model quality and inference operations are part of the product.
The brief
Context, requirements, acceptance criteria, non-goals, and the full production standard — as Markdown, ready for any coding agent.
Build brief — a focused alternative to Krisp
Context
**Krisp** — Real-time AI noise cancellation and meeting voice enhancement. It currently costs $16/mo.
You can build post-processing and maybe route audio through open models, but real-time low-latency virtual-device noise cancellation is not a one-sitting web app.
This brief describes a focused, single-operator replacement for the part of Krisp that is genuinely reproducible. It is deliberately narrower than the product it replaces, and it says so in writing. Build the useful core; do not pretend to have rebuilt the rest.
What you are building
Use an open noise-suppression model as a virtual microphone or post-process recordings before sending them to calls or files.
Build a focused single-user workflow with real persistence, search, and export.
A responsive interface with real empty, loading, success, and error states.
Requirements
Functional
Audio DSP library.
OS-level virtual audio device.
Model/runtime optimized for low latency.
MacOS/Windows audio permissions.
Non-functional
Accessibility: semantic markup, labelled controls, visible focus, and reduced-motion support.
Security: server-side secrets, validated input, and no credentials in the client bundle.
Reliability: retries with backoff on external calls, and a clear failure state when a provider is down.
Portability: the operator can export their data and leave without losing it.
Implementation brief
Build me a noise-cleanup tool to replace Krisp for recordings, not live calls. Requirements:
Be honest in the README up front: real-time suppression as a virtual microphone
needs an OS-level audio driver and millisecond latency. That is Krisp's actual
product and it is out of scope here. This tool cleans audio files after the fact.
A CLI: `denoise in.wav` (mp3/m4a accepted, decoded via ffmpeg) writes
in.clean.wav next to the original, original untouched.
Use RNNoise for the suppression pass (ffmpeg's arnndn filter with a downloaded
model file is the easy route), then a loudnorm pass so voice levels come out
consistent.
A watch mode with chokidar: drop files into ~/Denoise/in/ and cleaned versions
appear in ~/Denoise/out/ with the same names.
Batch mode for whole folders, with a printed before/after noise estimate per file.
Node wrapping ffmpeg, or a plain bash script if that ships simpler. No server,
no GUI.
Fully local and offline: no accounts, no telemetry, no API calls.
Out of scope: live call processing and a virtual audio device. For live calls the
README should point me at the OS's built-in voice isolation instead.
README: ffmpeg and RNNoise model install, one-line usage examples.
Delivery standard
Inspect the repository first, then write a short implementation plan before writing code.
Deliver the smallest complete end-to-end workflow first; every primary control must work against persisted data.
Use real validation and storage; never substitute fake dashboards, decorative controls, hard-coded success states, or mock integrations.
Include responsive layouts plus genuine empty, loading, success, validation, and failure states.
Keep secrets server-side in environment variables, provide .env.example, and never commit credentials or user data.
Add structured logs around every external call and return actionable errors without leaking sensitive details.
Write unit tests for the core logic and one automated test of the main user journey.
Finish with a README covering setup, architecture, data location, backups, tests, deployment, and known limitations.
Acceptance criteria
A clean install starts the app using only the README and .env.example.
The primary journey works from first visit through saved result, reload, edit, export, and deletion where applicable.
Invalid input, missing configuration, provider failure, and an empty database each have a usable state.
The interface works at 390px and 1440px, is keyboard navigable, and shows visible focus on every control.
Tests, type checking, linting, and a production build all pass with no ignored failures.
No part of the interface implies a live integration, security guarantee, or scale capability that was not actually built and verified.
Non-goals
Do not build these, and do not claim to have replaced them:
Low-latency virtual microphone.
Polished app switching.
Model quality.
Meeting integrations.
The last 20 percent is sync, migration fidelity, speed, and edge cases.
Model quality and inference operations are part of the product.
What you still own after launch
Run migrations, backups, restores, and dependency updates.
Test the critical journey after every model, API, or hosting change.
Monitor failures and fix the edge cases a first prompt will miss.
Risk
**Manageable.** A personal version is realistic if you test the critical journey and keep reliable backups.
Editorial confidence in this assessment: medium. No independent one-shot implementation is linked yet.
Prior art
Working open-source software you can read, fork, or borrow from before starting:
[RNNoise](https://github.com/xiph/rnnoise) — Open-source recurrent noise suppression model; useful but not a full Krisp replacement
Generated by [Can It Be Vibe Coded?](https://www.canitbevibecoded.com) · Full report: https://www.canitbevibecoded.com/krisp
You still own the product
- Run migrations, backups, restores, and dependency updates.
- Test the critical journey after every model, API, or hosting change.
- Monitor failures and fix the edge cases a first prompt will miss.
Open-source prior art
Before you start
Can Krisp be vibe coded?
Partly, if you narrow it. You can build post-processing and maybe route audio through open models, but real-time low-latency virtual-device noise cancellation is not a one-sitting web app.
What can an AI coding agent reproduce from Krisp?
Use an open noise-suppression model as a virtual microphone or post-process recordings before sending them to calls or files. Build a focused single-user workflow with real persistence, search, and export. A responsive interface with real empty, loading, success, and error states.
What will a DIY Krisp replacement still be missing?
low-latency virtual microphone; polished app switching; model quality; meeting integrations; The last 20 percent is sync, migration fidelity, speed, and edge cases.; Model quality and inference operations are part of the product.
What do I still own after building a Krisp alternative?
Run migrations, backups, restores, and dependency updates. Test the critical journey after every model, API, or hosting change. Monitor failures and fix the edge cases a first prompt will miss.