How to Use DeepFilterNet 3 in ChatGPT
DeepFilterNet 3 reduces background noise in recorded speech, making it useful for interviews, narration, and voice notes. It processes full-band audio at 48 kHz and brings slightly improved noise suppression over DeepFilterNet 2.
Prompts to try with DeepFilterNet 3 in ChatGPT
Copy an example into ChatGPT, or ask in your own words.
Use DeepFilterNet 3 to reduce background noise in this interview recording. Return a separate processed copy so I can compare it with the original.
Try DeepFilterNet 3 on a short passage of this podcast dialogue. Include a passage with quiet speech in the comparison.
Process this recorded lesson with DeepFilterNet 3. Keep the original available and return the processed narration for listening review.
Use DeepFilterNet 3 on this non-English voice note. Return the processed recording so I can check pronunciation and quiet words myself.
Try DeepFilterNet 3 on this narrowband radio excerpt. Return both versions for comparison before processing any more recordings.
How to set up DeepFilterNet 3 in ChatGPT
Pick the way that suits you. Each takes about two minutes.
Use the ChatGPT connector
- Add a custom MCP server
In ChatGPT open Plugins, then Add › Add custom MCP server. Name it Spiral, paste this URL and keep Authentication on OAuth. Tick I understand and want to continue, then click Create.
https://api.spiral.ai/mcp
- Click Connect
ChatGPT asks to connect Spiral. Connect takes you to Spiral in your browser.
- Sign in to Spiral and approve
Sign in, pick the organization ChatGPT should use and click Authorize. Spiral gives the connection its own API key; ChatGPT never sees it, and you can remove the connection under Settings › Connected apps.
Use the Spiral skill
- Send ChatGPT this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run DeepFilterNet 3
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "fal", "model": "fal-ai/deepfilternet3", "input": { "audio_url": "https://v3b.fal.media/files/b/0a8a4022/DLoZhabKeVjd3urncoRE4_dirty.mp3" } }'
Then paste one of the prompts above: ChatGPT finds DeepFilterNet 3 on Spiral, shows the price and runs it.
DeepFilterNet 3 providers and prices
Spiral runs DeepFilterNet 3 through fal. Open a card for today's price, with every fee included.
Settings you can ask for
DeepFilterNet 3 works from your audio URL. Ask ChatGPT for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Audio format | mp3, aac, m4a, ogg, opus, flac, or wav | mp3 | The format for the output audio. |
Every API field of DeepFilterNet 3 is on its tool page: open its card above.
What can you do with DeepFilterNet 3 in ChatGPT?
Create a noise-reduced listening copy of an interview, voice note, or spoken lesson. A chat agent can use DeepFilterNet 3 and help you organize a comparison with the original.
Podcast dialogue cleanup
Try background-noise reduction on spoken podcast tracks before editing. Keep the original available for checking quiet words.
Interview listening copies
Create processed copies of recorded interviews for editorial listening. Compare a representative passage before processing the remaining recordings.
Course narration
Reduce background noise in recorded lessons and spoken explanations. Review the result before adding music or assembling the final lesson.
Software demo narration
Process the spoken narration from a screencast before returning it to the video edit. Check the processed recording against the original.
Meeting recording review
Make a noise-reduced listening copy of recorded meeting speech. Retain the original when reviewing names, numbers, and quiet contributions.
Voice-note cleanup
Try speech enhancement on recorded voice notes before sharing a listening copy. Check that the words remain clear after processing.
More on DeepFilterNet 3 and ChatGPT
Overview
DeepFilterNet 3 in other agents
Questions about DeepFilterNet 3 in ChatGPT
Can ChatGPT use DeepFilterNet 3?
Yes. Add Spiral to ChatGPT as an MCP server, and ChatGPT can find and run DeepFilterNet 3 like any built-in tool.
How much does DeepFilterNet 3 cost in ChatGPT?
Up to $0.078 per run at the default inputs for 60 seconds of input. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your ChatGPT plan is separate.
Do I need a fal account to use DeepFilterNet 3 in ChatGPT?
No. Spiral runs DeepFilterNet 3 on its own fal account. You sign in to Spiral once and never paste a key.
How long does it take to set up DeepFilterNet 3 in ChatGPT?
About two minutes: add a custom MCP server, then click Connect, then sign in to Spiral and approve.
Does ChatGPT wait while DeepFilterNet 3 runs?
DeepFilterNet 3 runs in the background. ChatGPT starts the run, checks on it and gives you the audio when it's ready.
What is DeepFilterNet 3 useful for?
DeepFilterNet 3 reduces background noise in speech. Suggested uses include spoken interviews, podcast dialogue, course narration, and meeting recordings.
Does DeepFilterNet 3 work with languages other than English?
The research paper describes training on multilingual DNS4 data. That supports trying it on non-English speech, but does not establish equal quality for every language or accent.
What audio sample rate does DeepFilterNet 3 use?
Its native processing rate is 48 kHz. This describes the speech-enhancement model, rather than a guarantee about the sample rate of every exported recording.
How does DeepFilterNet 3 differ from DeepFilterNet 2?
The maker describes slightly improved noise suppression. Treat it as an incremental improvement and compare results on your own recordings.
Can DeepFilterNet 3 damage quiet speech?
An upstream user reported weakened speech alongside residual noise when processing narrowband radio recordings. That report is a reason to check quiet words against the original, especially with radio audio.
Start using DeepFilterNet 3 in ChatGPT
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into ChatGPT:
set up https://spiral.ai/SKILL.md


