Denoise (DeepFilterNet)

Remove background noise from audio using DeepFilterNet 3, a state-of-the-art ML-based denoiser. Preserves natural voice timbre better than rnnoise or ffmpeg's afftdn — ideal for podcast, interview, and voice-memo cleanup.

When to use

• The user has a noisy podcast recording or interview.
• The source is speech-heavy (podcast, voice memo, audiobook); DeepFilterNet is tuned for voice.
• Existing denoise (rnnoise, afftdn) leaves artifacts (hollowness, over-suppression) → try DeepFilterNet for comparison.
• The user wants a single-pass, set-and-forget denoise without threshold tweaking.

Do not use this skill for:

• Music with noise — DeepFilterNet is speech-optimized; may remove subtle instruments.
• Surgical noise reduction needing manual control → use ffmpeg afftdn or rnnoise with threshold tuning.

Inputs