From audio-production
Use when the user wants to denoise audio using modern ML-based filtering. DeepFilterNet produces cleaner speech than traditional FFT/rnnoise methods — no watery/hollow artifacts. Wraps `deepFilter` binary for podcast/voice cleanup.
How this skill is triggered — by the user, by Claude, or both
Slash command
/audio-production:denoise-deepfilterThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
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.
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.
Do not use this skill for:
afftdn or rnnoise with threshold tuning.<input-stem>.deepfilter.<ext>.1–40 dB. Lower = more aggressive. Common: 10–20 dB for heavy noise.deepfilter-v3 or deepfilter-v2 if needed for compatibility.Check if deepFilter is on PATH (which deepFilter). If missing, provide install instructions:
https://github.com/Rikorose/DeepFilterNet/releases. Unzip and add to PATH.cargo install deep_filter.install-deps if the plugin includes it.Build the command:
deepFilter \
"<input-file>" \
-o "<output-file>" \
[--attenuation-limit <dB>]
If --attenuation-limit is provided, insert it. Example: --attenuation-limit 15.
Run the command and capture stdout/stderr. DeepFilterNet reports progress (model loading, processing).
Verify output: ffprobe -i "<output-file>" -hide_banner to confirm the file is valid and report duration (should match input).
<input> → <output> (denoised, attenuation-limit <limit|unlimited>, duration <N>s).ffmpeg -i <output> -filter:a "loudnorm" to re-normalize.--attenuation-limit or trying denoise-rnnoise as an alternative.Install deepFilter binary from GitHub releases or via cargo install deep_filter. Requires no Python or special runtime — it's a standalone Rust binary. Audio format support via ffmpeg or embedded codecs.
Searches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.
Guides Payload CMS config (payload.config.ts), collections, fields, hooks, access control, APIs. Debugs validation errors, security, relationships, queries, transactions, hook behavior.
Implements vector databases with Pinecone, Weaviate, Qdrant, Milvus, pgvector for semantic search, RAG, recommendations, and similarity systems. Optimizes embeddings, indexing, and hybrid search.
npx claudepluginhub danielrosehill/claude-code-plugins --plugin audio-production