[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AS4XS86Y6EFK7GNCY90DFB":3,"skill-same:01M7AS4XS86Y6EFK7GNCY90DFB":48},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M7AS4XS86Y6EFK7GNCY90DFB","dsp-audio-rf-and-images",null,"Use when working in a specific signal domain: audio and speech (psychoacoustics, coding and codecs, the real-time voice pipeline with echo cancellation, noise suppression and latency budgets, music and analysis), RF and communications (modulation, software-defined radio, synchronization, equalization), and images and video treated as two-dimensional signals.","the-vibey-project\u002Fvibey","the-vibey-project","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F329298758?v=4","01M7AJHGZ9722CTJ4JNQZVRH0Q","the vibey project","01M7AJHGZVMG2K7TGXQ6XFHVG3",5,"MIT","https:\u002F\u002Fgithub.com\u002Fthe-vibey-project\u002Fvibey\u002Fblob\u002Fdevelop\u002Fsrc\u002Fvibey_tools\u002Fskills\u002Fplugins\u002Fsignal-processing\u002Fskills\u002Fdsp-audio-rf-and-images\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: dsp-audio-rf-and-images\ndescription: \"Use when working in a specific signal domain: audio and speech (psychoacoustics, coding and codecs, the real-time voice pipeline with echo cancellation, noise suppression and latency budgets, music and analysis), RF and communications (modulation, software-defined radio, synchronization, equalization), and images and video treated as two-dimensional signals.\"\n---\n\n# Signal Processing: Audio and Speech, RF and Communications, and Images and Video\n\n> **Part 3 of 5** of the *Signal Processing* reference (plugin `signal-processing`), covering §8–§10. Sibling skills: `dsp-sampling-frequency-domain-and-filters` (§0–§3), `dsp-convolution-multirate-and-spectral-analysis` (§4–§7), `dsp-implementation-tools-and-testing` (§11–§14), `dsp-reference` (§15–§20). Section numbers are shared across the set; a reference written as §N → `skill` points into that sibling skill.\n>\n> **Currency:** Verified August 2026. See §17 → `dsp-reference` for the currency snapshot and what goes stale first.\n\n> **How to read this.** Reference for engineers who need DSP working, not a course. Three\n> markers:\n> - **[DURABLE]** — sampling theory, transforms, filters, estimation. **Nyquist is 1928;\n>   Shannon 1949; the Cooley–Tukey FFT 1965. This material does not expire.**\n> - **[VERSIONED]** — codecs, libraries, hardware, the ML layer.\n> - **[CONTESTED]** — genuine disagreement.\n>\n> **⚠️ GOTCHA** boxes mark where the code runs, the plot looks fine, and the answer is\n> wrong.\n>\n> **The three framings that organize everything below:**\n> 1. **Everything is a trade between time and frequency resolution, and you cannot cheat\n>    it.** The uncertainty principle is not a metaphor here — **a short window gives you\n>    good time resolution and poor frequency resolution, and no amount of cleverness\n>    escapes that** (§2.4 → `dsp-sampling-frequency-domain-and-filters`). Most DSP design is choosing where on that curve to sit.\n> 2. **⚠️ Aliasing is irreversible.** Once you've sampled too slowly, the information is\n>    gone — no filter, no algorithm, and no neural network recovers it. **This is the one\n>    mistake in this document that cannot be fixed downstream** (§1 → `dsp-sampling-frequency-domain-and-filters`).\n> 3. **The field is mid-convergence with machine learning, and the interesting result is\n>    that hybrids are winning** — not pure DSP, and not pure end-to-end learning.\n>    ⚠️ **The pattern that keeps recurring: use DSP for the structure you know, and learn\n>    the parameters you can't hand-tune** (§8, §16.1 → `dsp-reference`).\n\n---\n\n## §8. Audio and Speech\n\n### 8.1 The domain facts\n\n**[DURABLE]** Human hearing spans roughly **20 Hz – 20 kHz** (⚠️ **and the top end\ndeclines substantially with age**), with **~120 dB of dynamic range**, **logarithmic**\nin both frequency and amplitude. **Critical bands \u002F Bark \u002F ERB scales** model the ear's\nfrequency resolution, and **masking** — a loud tone hiding a nearby quiet one, in frequency\nand in time — is the entire basis of perceptual audio coding.\n\n⚠️ **Latency budgets are what constrain audio engineering**: musicians notice round-trip\nlatency above roughly **10 ms**; conversational latency becomes awkward above ~**150–200 ms**\none-way. **These numbers determine your block size, and block size determines everything\nelse.**\n\n### 8.2 Coding\n\n**Perceptual codecs** (MP3, AAC, Vorbis, Opus) throw away what masking says you can't\nhear: **MDCT** transform, psychoacoustic model, quantize, entropy-code.\n**Speech codecs** (⚠️ **model the vocal tract rather than the waveform**): LPC, CELP,\nAMR, and Opus's SILK layer.\n\n**[VERSIONED] Opus is the one to know** — RFC 6716, hybrid SILK + CELT, **mandatory in\nWebRTC**, and covers everything from speech at low bitrate to full-band music.\n\n> **⚠️ GOTCHA — and the most interesting story in DSP right now: classical codecs are\n> absorbing neural components rather than being replaced by them.**\n>\n> **Opus 1.5 (March 2024) put machine learning inside the codec for the first time**, all\n> at the decoder — meaning **a service can adopt it on the playback side alone** and\n> improve calls for users on bad networks without changing the sender.\n> - **Deep PLC** — a neural network reconstructs lost packets, in place of the classical\n>   repeat-and-fade. Recommended for occasional loss.\n> - **DRED (Deep REDundancy)** — for **burst** loss. It uses a **rate-distortion-optimized\n>   VAE** to compress acoustic features so aggressively that **up to one second of\n>   redundancy fits in ~12–32 kb\u002Fs of overhead**, carried in each packet's padding.\n>   ⚠️ **It transmits 20 acoustic features — 18 Bark-frequency cepstral coefficients plus\n>   pitch and voicing — and resynthesizes speech from them.** Note the consequence: **the\n>   features contain no phase information, so the recovered waveform deviates\n>   significantly from the original while sounding similar.**\n> - **LACE\u002FNoLACE** — a DNN post-filter that denoises the decoder's output.\n>\n> **Opus 1.6 (Dec 2025)** added **bandwidth extension (BWE)**, which combined with DRED\n> reaches **fullband quality**, and with NoLACE enables **good fullband speech as low as\n> 9 kb\u002Fs.**\n>\n> ⚠️ **Two caveats worth carrying.** **DRED's format was still being finalised in IETF as\n> of 2026 — treat it as advanced rather than settled.** And **any DNN in a codec\n> significantly increases real-time compute**, which is the actual adoption barrier —\n> along with model size, which Opus's own developers name as a main obstacle.\n\n**Neural codecs proper** — SoundStream, EnCodec, and the discrete-token codecs underpinning\ngenerative audio — are a genuinely different lineage: much lower bitrates, much higher\ncompute, and **tokens that double as inputs to generative models.**\n\n### 8.3 The real-time voice pipeline\n\n**[DURABLE] The order matters and is roughly fixed**: **AEC → noise suppression →\ndereverberation → AGC → codec.**\n\n**Acoustic echo cancellation** — ⚠️ **adaptive filter (§6 → `dsp-convolution-multirate-and-spectral-analysis`) to model the echo path, plus a\nnonlinear residual echo suppressor, plus a double-talk detector.** Interestingly, **most\nstate-of-the-art AEC is still classical DSP or hybrid DSP-ML**, with deep learning\ntypically replacing only the nonlinear residual suppressor.\n\n**Noise suppression — and this is where the convergence story is clearest.** The classical\nline runs **spectral subtraction (Boll)** → ⚠️ **which produced the notorious \"musical\nnoise\" artifact** → **Wiener filtering** (needs a priori statistics) → **MMSE log-spectral\namplitude (Ephraim–Malah, 1985)**, still a respectable baseline.\n\n**RNNoise (Valin, 2018) is the landmark hybrid**: a small recurrent network estimates\n**ideal critical-band gains** while **a traditional pitch filter attenuates noise between\nharmonics** — ⚠️ **significantly better than a classical MMSE estimator while staying\nreal-time at 48 kHz on a low-power CPU.** **DeepFilterNet** and successors extended this by\n**predicting linear filters rather than estimating clean speech directly**, which is\ncomputationally far cheaper and edge-deployable.\n\n**⚠️ The honest current comparison**: traditional approaches are **fast but conservative**;\ndeep learning can be **more aggressive while preserving speech quality, at higher\ncomputational cost.** In hearing-aid studies, deep and classical methods perform\n**similarly in diffuse noise**, with the deep binaural approach winning **in the presence\nof spatial interferers.** **The pattern throughout: hybrids, not replacements.**\n\n**Beamforming** for multi-mic arrays — delay-and-sum, MVDR, GSC — ⚠️ **spatial filtering\nthat is genuinely complementary to spectral methods**, and increasingly paired with neural\nmask estimation.\n\n### 8.4 Music and analysis\n**Pitch detection** (autocorrelation, YIN, CREPE), **onset detection**, **beat tracking**,\n**source separation** (⚠️ **Demucs, Spleeter and the neural separators substantially beat\nclassical NMF-based methods** — one of the cleanest ML wins in audio), **MFCCs** (⚠️ **the\nclassical speech feature; largely superseded by learned representations and mel\nspectrograms in modern ML pipelines, but still everywhere in legacy code**), **time-scale\nand pitch modification** (phase vocoder, PSOLA, WSOLA).\n\n---\n\n## §9. RF and Communications\n\n**Modulation**: analogue (AM\u002FFM\u002FPM), digital (**ASK\u002FFSK\u002FPSK\u002FQAM**), and **OFDM** —\n⚠️ **which converts a frequency-selective channel into many flat sub-channels and is why\nit's in Wi-Fi, LTE, 5G, and DVB.** **Spread spectrum** (DSSS, FHSS) for interference\nresistance and multiple access.\n\n**The receive chain**: down-conversion to **I\u002FQ baseband** (⚠️ **complex-valued\nrepresentation is the whole language of RF DSP — get comfortable with it**),\n**matched filtering** (§4 → `dsp-convolution-multirate-and-spectral-analysis`), **timing and carrier recovery**, **equalization** (§6 → `dsp-convolution-multirate-and-spectral-analysis`), and\n**demodulation**.\n\n**Error correction**: Hamming and Reed–Solomon (⚠️ **still the standard for burst errors in\nstorage and broadcast**), convolutional codes with Viterbi decoding, **Turbo**, **LDPC**\n(⚠️ **near-Shannon-limit, and in 5G, Wi-Fi 6 and DVB-S2**), and **polar codes** in 5G\ncontrol channels.\n\n**⚠️ The physical realities that dominate**: multipath and fading, Doppler, the noise\nfigure of your front end, **and the fact that Shannon capacity is a hard ceiling** —\nC = B log₂(1 + SNR). **No modulation scheme beats it.**\n\n**SDR** is where software engineers actually meet this: **GNU Radio** (flowgraph-based),\n**RTL-SDR** (⚠️ **~$30 and genuinely capable for receive-only experimentation**),\n**HackRF**, **USRP**, **LimeSDR**, and **SoapySDR** as the hardware abstraction.\n⚠️ **Transmitting requires a licence in most jurisdictions and most bands — receiving\ngenerally doesn't.**\n\n---\n\n## §10. Images and Video\n\n**[DURABLE] 2D signal processing, and the concepts transfer directly.**\n**2D convolution** for blur, sharpen, and edge detection (Sobel, Laplacian);\n**separable kernels** (⚠️ **a 2D Gaussian is two 1D passes — O(n) instead of O(n²) per\npixel, and this is why Gaussian blur is cheap**); **2D FFT** for frequency-domain\nfiltering; **the DCT** as JPEG's core.\n\n**⚠️ Aliasing appears here as moiré patterns** — and image downsampling without\npre-filtering is exactly §1.1 → `dsp-sampling-frequency-domain-and-filters`'s error in two dimensions. **Anti-aliasing in rendering is\nthe same problem approached from the synthesis side.**\n\n**Also**: **image pyramids** (Gaussian, Laplacian) for multi-scale processing;\n**bilateral and non-local means** filters (⚠️ **edge-preserving — linear filters can't do\nthis**); **optical flow** for motion; **video codecs** as motion-compensated transform\ncoding (H.264\u002FAVC, H.265\u002FHEVC, AV1, and H.266\u002FVVC).\n","# Signal Processing: Audio and Speech, RF and Communications, and Images and Video\n\n> **Part 3 of 5** of the *Signal Processing* reference (plugin `signal-processing`), covering §8–§10. Sibling skills: `dsp-sampling-frequency-domain-and-filters` (§0–§3), `dsp-convolution-multirate-and-spectral-analysis` (§4–§7), `dsp-implementation-tools-and-testing` (§11–§14), `dsp-reference` (§15–§20). Section numbers are shared across the set; a reference written as §N → `skill` points into that sibling skill.\n>\n> **Currency:** Verified August 2026. See §17 → `dsp-reference` for the currency snapshot and what goes stale first.\n\n> **How to read this.** Reference for engineers who need DSP working, not a course. Three\n> markers:\n> - **[DURABLE]** — sampling theory, transforms, filters, estimation. **Nyquist is 1928;\n>   Shannon 1949; the Cooley–Tukey FFT 1965. This material does not expire.**\n> - **[VERSIONED]** — codecs, libraries, hardware, the ML layer.\n> - **[CONTESTED]** — genuine disagreement.\n>\n> **⚠️ GOTCHA** boxes mark where the code runs, the plot looks fine, and the answer is\n> wrong.\n>\n> **The three framings that organize everything below:**\n> 1. **Everything is a trade between time and frequency resolution, and you cannot cheat\n>    it.** The uncertainty principle is not a metaphor here — **a short window gives you\n>    good time resolution and poor frequency resolution, and no amount of cleverness\n>    escapes that** (§2.4 → `dsp-sampling-frequency-domain-and-filters`). Most DSP design is choosing where on that curve to sit.\n> 2. **⚠️ Aliasing is irreversible.** Once you've sampled too slowly, the information is\n>    gone — no filter, no algorithm, and no neural network recovers it. **This is the one\n>    mistake in this document that cannot be fixed downstream** (§1 → `dsp-sampling-frequency-domain-and-filters`).\n> 3. **The field is mid-convergence with machine learning, and the interesting result is\n>    that hybrids are winning** — not pure DSP, and not pure end-to-end learning.\n>    ⚠️ **The pattern that keeps recurring: use DSP for the structure you know, and learn\n>    the parameters you can't hand-tune** (§8, §16.1 → `dsp-reference`).\n\n---\n\n## §8. Audio and Speech\n\n### 8.1 The domain facts\n\n**[DURABLE]** Human hearing spans roughly **20 Hz – 20 kHz** (⚠️ **and the top end\ndeclines substantially with age**), with **~120 dB of dynamic range**, **logarithmic**\nin both frequency and amplitude. **Critical bands \u002F Bark \u002F ERB scales** model the ear's\nfrequency resolution, and **masking** — a loud tone hiding a nearby quiet one, in frequency\nand in time — is the entire basis of perceptual audio coding.\n\n⚠️ **Latency budgets are what constrain audio engineering**: musicians notice round-trip\nlatency above roughly **10 ms**; conversational latency becomes awkward above ~**150–200 ms**\none-way. **These numbers determine your block size, and block size determines everything\nelse.**\n\n### 8.2 Coding\n\n**Perceptual codecs** (MP3, AAC, Vorbis, Opus) throw away what masking says you can't\nhear: **MDCT** transform, psychoacoustic model, quantize, entropy-code.\n**Speech codecs** (⚠️ **model the vocal tract rather than the waveform**): LPC, CELP,\nAMR, and Opus's SILK layer.\n\n**[VERSIONED] Opus is the one to know** — RFC 6716, hybrid SILK + CELT, **mandatory in\nWebRTC**, and covers everything from speech at low bitrate to full-band music.\n\n> **⚠️ GOTCHA — and the most interesting story in DSP right now: classical codecs are\n> absorbing neural components rather than being replaced by them.**\n>\n> **Opus 1.5 (March 2024) put machine learning inside the codec for the first time**, all\n> at the decoder — meaning **a service can adopt it on the playback side alone** and\n> improve calls for users on bad networks without changing the sender.\n> - **Deep PLC** — a neural network reconstructs lost packets, in place of the classical\n>   repeat-and-fade. Recommended for occasional loss.\n> - **DRED (Deep REDundancy)** — for **burst** loss. It uses a **rate-distortion-optimized\n>   VAE** to compress acoustic features so aggressively that **up to one second of\n>   redundancy fits in ~12–32 kb\u002Fs of overhead**, carried in each packet's padding.\n>   ⚠️ **It transmits 20 acoustic features — 18 Bark-frequency cepstral coefficients plus\n>   pitch and voicing — and resynthesizes speech from them.** Note the consequence: **the\n>   features contain no phase information, so the recovered waveform deviates\n>   significantly from the original while sounding similar.**\n> - **LACE\u002FNoLACE** — a DNN post-filter that denoises the decoder's output.\n>\n> **Opus 1.6 (Dec 2025)** added **bandwidth extension (BWE)**, which combined with DRED\n> reaches **fullband quality**, and with NoLACE enables **good fullband speech as low as\n> 9 kb\u002Fs.**\n>\n> ⚠️ **Two caveats worth carrying.** **DRED's format was still being finalised in IETF as\n> of 2026 — treat it as advanced rather than settled.** And **any DNN in a codec\n> significantly increases real-time compute**, which is the actual adoption barrier —\n> along with model size, which Opus's own developers name as a main obstacle.\n\n**Neural codecs proper** — SoundStream, EnCodec, and the discrete-token codecs underpinning\ngenerative audio — are a genuinely different lineage: much lower bitrates, much higher\ncompute, and **tokens that double as inputs to generative models.**\n\n### 8.3 The real-time voice pipeline\n\n**[DURABLE] The order matters and is roughly fixed**: **AEC → noise suppression →\ndereverberation → AGC → codec.**\n\n**Acoustic echo cancellation** — ⚠️ **adaptive filter (§6 → `dsp-convolution-multirate-and-spectral-analysis`) to model the echo path, plus a\nnonlinear residual echo suppressor, plus a double-talk detector.** Interestingly, **most\nstate-of-the-art AEC is still classical DSP or hybrid DSP-ML**, with deep learning\ntypically replacing only the nonlinear residual suppressor.\n\n**Noise suppression — and this is where the convergence story is clearest.** The classical\nline runs **spectral subtraction (Boll)** → ⚠️ **which produced the notorious \"musical\nnoise\" artifact** → **Wiener filtering** (needs a priori statistics) → **MMSE log-spectral\namplitude (Ephraim–Malah, 1985)**, still a respectable baseline.\n\n**RNNoise (Valin, 2018) is the landmark hybrid**: a small recurrent network estimates\n**ideal critical-band gains** while **a traditional pitch filter attenuates noise between\nharmonics** — ⚠️ **significantly better than a classical MMSE estimator while staying\nreal-time at 48 kHz on a low-power CPU.** **DeepFilterNet** and successors extended this by\n**predicting linear filters rather than estimating clean speech directly**, which is\ncomputationally far cheaper and edge-deployable.\n\n**⚠️ The honest current comparison**: traditional approaches are **fast but conservative**;\ndeep learning can be **more aggressive while preserving speech quality, at higher\ncomputational cost.** In hearing-aid studies, deep and classical methods perform\n**similarly in diffuse noise**, with the deep binaural approach winning **in the presence\nof spatial interferers.** **The pattern throughout: hybrids, not replacements.**\n\n**Beamforming** for multi-mic arrays — delay-and-sum, MVDR, GSC — ⚠️ **spatial filtering\nthat is genuinely complementary to spectral methods**, and increasingly paired with neural\nmask estimation.\n\n### 8.4 Music and analysis\n**Pitch detection** (autocorrelation, YIN, CREPE), **onset detection**, **beat tracking**,\n**source separation** (⚠️ **Demucs, Spleeter and the neural separators substantially beat\nclassical NMF-based methods** — one of the cleanest ML wins in audio), **MFCCs** (⚠️ **the\nclassical speech feature; largely superseded by learned representations and mel\nspectrograms in modern ML pipelines, but still everywhere in legacy code**), **time-scale\nand pitch modification** (phase vocoder, PSOLA, WSOLA).\n\n---\n\n## §9. RF and Communications\n\n**Modulation**: analogue (AM\u002FFM\u002FPM), digital (**ASK\u002FFSK\u002FPSK\u002FQAM**), and **OFDM** —\n⚠️ **which converts a frequency-selective channel into many flat sub-channels and is why\nit's in Wi-Fi, LTE, 5G, and DVB.** **Spread spectrum** (DSSS, FHSS) for interference\nresistance and multiple access.\n\n**The receive chain**: down-conversion to **I\u002FQ baseband** (⚠️ **complex-valued\nrepresentation is the whole language of RF DSP — get comfortable with it**),\n**matched filtering** (§4 → `dsp-convolution-multirate-and-spectral-analysis`), **timing and carrier recovery**, **equalization** (§6 → `dsp-convolution-multirate-and-spectral-analysis`), and\n**demodulation**.\n\n**Error correction**: Hamming and Reed–Solomon (⚠️ **still the standard for burst errors in\nstorage and broadcast**), convolutional codes with Viterbi decoding, **Turbo**, **LDPC**\n(⚠️ **near-Shannon-limit, and in 5G, Wi-Fi 6 and DVB-S2**), and **polar codes** in 5G\ncontrol channels.\n\n**⚠️ The physical realities that dominate**: multipath and fading, Doppler, the noise\nfigure of your front end, **and the fact that Shannon capacity is a hard ceiling** —\nC = B log₂(1 + SNR). **No modulation scheme beats it.**\n\n**SDR** is where software engineers actually meet this: **GNU Radio** (flowgraph-based),\n**RTL-SDR** (⚠️ **~$30 and genuinely capable for receive-only experimentation**),\n**HackRF**, **USRP**, **LimeSDR**, and **SoapySDR** as the hardware abstraction.\n⚠️ **Transmitting requires a licence in most jurisdictions and most bands — receiving\ngenerally doesn't.**\n\n---\n\n## §10. Images and Video\n\n**[DURABLE] 2D signal processing, and the concepts transfer directly.**\n**2D convolution** for blur, sharpen, and edge detection (Sobel, Laplacian);\n**separable kernels** (⚠️ **a 2D Gaussian is two 1D passes — O(n) instead of O(n²) per\npixel, and this is why Gaussian blur is cheap**); **2D FFT** for frequency-domain\nfiltering; **the DCT** as JPEG's core.\n\n**⚠️ Aliasing appears here as moiré patterns** — and image downsampling without\npre-filtering is exactly §1.1 → `dsp-sampling-frequency-domain-and-filters`'s error in two dimensions. **Anti-aliasing in rendering is\nthe same problem approached from the synthesis side.**\n\n**Also**: **image pyramids** (Gaussian, Laplacian) for multi-scale processing;\n**bilateral and non-local means** filters (⚠️ **edge-preserving — linear filters can't do\nthis**); **optical flow** for motion; **video codecs** as motion-compensated transform\ncoding (H.264\u002FAVC, H.265\u002FHEVC, AV1, and H.266\u002FVVC).\n",{"name":5,"description":7},"develop","ae32736a1aabddabca5d2c96d665aa0ebc7d8700","src\u002Fvibey_tools\u002Fskills\u002Fplugins\u002Fsignal-processing\u002Fskills\u002Fdsp-audio-rf-and-images",3,"https:\u002F\u002Fgithub.com\u002Fthe-vibey-project\u002Fvibey","https:\u002F\u002Fthe-vibey-project.github.io\u002Fvibey\u002F","https:\u002F\u002Fgithub.com\u002Fthe-vibey-project","https:\u002F\u002Fvibewithadam.matthewsteinberger.com","freedom for software engineers","2026-11-15 05:47:42",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add the-vibey-project\u002Fvibey --skill dsp-audio-rf-and-images",{"tool":40,"command":41},"bunx","bunx skills add the-vibey-project\u002Fvibey --skill dsp-audio-rf-and-images",{"tool":43,"command":44},"pnpm","pnpm dlx skills add the-vibey-project\u002Fvibey --skill dsp-audio-rf-and-images","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fthe-vibey-project\u002Fvibey\u002Fblob\u002Fdevelop\u002Fsrc\u002Fvibey_tools\u002Fskills\u002Fplugins\u002Fsignal-processing\u002Fskills\u002Fdsp-audio-rf-and-images\u002FSKILL.md",true,[],[49,56,63,70,77,84],{"skillNo":50,"name":51,"nameZh":6,"description":52,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":53,"categories":54,"keywords":55,"publishedAt":19},"01M7AT1FQJAE1FQYAMDTF490Y4","white-paper-writing","Use when writing white papers, technical reports, policy documents, thought leadership pieces, or any long-form authoritative document. Triggers on requests to write, structure, outline, or review white papers, technical reports, position papers, policy briefs, or evidence-driven arguments. Also use for B2B content marketing strategy, technical documentation requiring authority, academic-style reports, or any document that must educate while persuading. Use when someone asks \"how should I structure this\" for a long-form document.","https:\u002F\u002Fgithub.com\u002Fthe-vibey-project\u002Fvibey\u002Fblob\u002Fdevelop\u002Fsrc\u002Fvibey_tools\u002Fskills\u002Fplugins\u002Fwriting-craft\u002Fskills\u002Fwhite-paper-writing\u002FSKILL.md",[],[],{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7AT16BJRJJKK2MASXRZS76G","prose-technical-mechanics","Use when editing or writing prose at a technical\u002Fmechanical level: choosing between Anglo-Saxon and Latinate diction, placing information in topic vs stress positions, writing periodic vs cumulative sentences, handling sentence rhythm, punctuation as logic, transitions, paragraph structure, or phonesthetics. Use when someone wants to understand WHY prose works or doesn't, not just surface-level style advice. Triggers on word choice, sentence structure, emphasis, rhythm, diction, or any deep writing mechanics question.","https:\u002F\u002Fgithub.com\u002Fthe-vibey-project\u002Fvibey\u002Fblob\u002Fdevelop\u002Fsrc\u002Fvibey_tools\u002Fskills\u002Fplugins\u002Fwriting-craft\u002Fskills\u002Fprose-technical-mechanics\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7AT0W9HT0ZZ9Y287NW99MR3","prose-mechanics","Use when writing, editing, or analyzing prose style; when trying to avoid generic AI-sounding writing; when applying specific prose styles such as journalistic, academic, narrative, KJV biblical, Hemingway plain, Latinate\u002FJamesian, Orwellian, Augustan, Victorian, Early Modern (Bacon, Browne, Milton), or free indirect discourse; or when coaching someone on writing voice and mechanics. Triggers on any request to improve, analyze, or write in a specific prose style, to fix \"AI-sounding\" writing, to make prose more vivid or concrete, to vary sentence rhythm, or to understand how a particular writer achieves their effects.","https:\u002F\u002Fgithub.com\u002Fthe-vibey-project\u002Fvibey\u002Fblob\u002Fdevelop\u002Fsrc\u002Fvibey_tools\u002Fskills\u002Fplugins\u002Fwriting-craft\u002Fskills\u002Fprose-mechanics\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7AT0KR635JX66X2X6F3E2CP","narrative-structure","Use when structuring stories, pitches, presentations, white papers, or any persuasive narrative; analyzing the hero's journey in fiction or non-fiction; applying StoryBrand for business communication; designing narrative arcs for thought leadership; or understanding Vogler, Snyder, Propp, or Campbell frameworks. Triggers on any narrative structure, storytelling, pitch structure, StoryBrand, hero's journey, three-act structure, Freytag, Save the Cat, narrative arc, or \"how should I structure this story\u002Fpitch\u002Fpresentation\" request.","https:\u002F\u002Fgithub.com\u002Fthe-vibey-project\u002Fvibey\u002Fblob\u002Fdevelop\u002Fsrc\u002Fvibey_tools\u002Fskills\u002Fplugins\u002Fwriting-craft\u002Fskills\u002Fnarrative-structure\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7AT0AKGW09NKKZDYBT8VRDE","legal-brief-writing","Use when writing legal briefs, motions, or any court filing; when analyzing judicial cognition and persuasion; when applying elite brief-writing techniques from Roberts, Clement, Kagan, Garner; or when structuring legal arguments for maximum persuasion. Also use for autistic\u002Fneurodivergent advocates who need to understand how their natural communication style maps to elite legal advocacy. Triggers on any legal writing, brief drafting, motion preparation, or legal argument structuring request.","https:\u002F\u002Fgithub.com\u002Fthe-vibey-project\u002Fvibey\u002Fblob\u002Fdevelop\u002Fsrc\u002Fvibey_tools\u002Fskills\u002Fplugins\u002Fwriting-craft\u002Fskills\u002Flegal-brief-writing\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7AT00VWPS4Q8B5VEMXX9NQT","journal-publishability-criteria","Use whenever a research paper, article or preprint is being written, revised, reviewed or published, and specifically before every publication of docs\u002Fpaper.md. Covers what legitimate journals actually gate on, in order of gating — the editor's desk screen (scope, novelty, package problems), what reviewers judge (methodological soundness, claims the evidence can carry, statistical and qualitative reporting), EQUATOR reporting guidelines, the ICMJE authorship criteria, COPE's view of text recycling, the near-universal generative-AI policy (no AI author, disclosure, author responsibility), TOP 2025 transparency standards, how to verify a journal is legitimate, submitting without an institution, and a pre-submission checklist. Triggers on requests to evaluate, review, prepare or submit a paper, to judge publishability, to write a declarations or data-availability section, to disclose AI use, or to check a journal.","https:\u002F\u002Fgithub.com\u002Fthe-vibey-project\u002Fvibey\u002Fblob\u002Fdevelop\u002Fsrc\u002Fvibey_tools\u002Fskills\u002Fplugins\u002Fwriting-craft\u002Fskills\u002Fjournal-publishability-criteria\u002FSKILL.md",[],[]]