Daniel Rosehill

My Weird Prompts

The human-AI collaboration podcast. A man, a sloth, and a donkey collaborate to create a podcast (with a little help from AI). No question is too obscure, no rabbit hole too deep. My Weird Prompts celebrates curiosity in all its forms. Daniel, the human, asks the questions that pop into his head at inconvenient moments. Corn the Sloth offers laid-back, thoughtful takes. Herman the Donkey brings boundless enthusiasm and energy. Together, they explore topics ranging from the mundane to the mind-bending. Each episode begins with a real voice memo from Daniel, processed through an AI pipeline that...

Vizitează neapărat site-ul podcastului și susține-i creatorul: myweirdprompts.com

Autor

Daniel Rosehill

Categorie

Technology

Site-ul podcastului

myweirdprompts.com

Cel mai nou episod

21 nov. 2026

Unde asculți?

Podcasturi în aplicație Replaio Radio În curând

Podcasturile ajung în curând în aplicație. Instaleaz-o acum și fii primul care descoperă o abordare complet nouă a podcasturilor

Descarcă din Google Play Instalează gratuit Android 10 mil.+ descărcări · nota 4,7+ iOS în curând

Episoade

Controlling a Web App While Walking 21.11.2026

Controlling a custom web app with a phone tucked away in a pocket runs headfirst into Android's input architecture. While tiny Bluetooth remotes and voice commands seem appealing for speech recording on the go, Chromium drops keyboard events the instant a page loses visibility. This breakdown explores why media session keys succeed where handheld gamepads fail, the drawbacks of voice triggers in s...

Why Weight Initialisation Depends on Architecture, Not Task 21.11.2026

Random weight initialization is not a one-size-fits-all roll of the dice dictated by the machine learning task. The math behind starting weights depends entirely on activation functions, normalization, and network depth. From classic LeCun and He formulas to orthogonal matrices, data-dependent scaling, and deterministic biases, exploring how models like nanoGPT balance signal variance shows why ar...

Recording Clean Voice Datasets While Walking 21.11.2026

Speech recognition datasets easily tolerate digital signal processing, but text-to-speech voice models memorize every limiter, noise gate, and ducked volume level as part of a speaker's actual voice. Recording clean, dual-purpose training data while pacing around means rethinking equipment from scratch. A close look at body-mounted dynamic mics, headworn options like the Shure WH20XLR, and handhel...

Why Universal Tripod Mounts Fail on Walls 21.11.2026

The quarter-inch-twenty UNC thread remains the standard tripod connection on modern cameras despite metric construction everywhere else. But mounting screens and cameras to walls exposes major hardware blind spots: shallow sockets that bow under long screws, aluminium that strips under five times normal grip torque, and unstandardized arm ratings. This look at small-rig mechanics breaks down why c...

Why Soundproof Curtains Only Work as Absorbers 21.11.2026

Soundproof curtains promise quiet, but the physics of sound isolation tell a very different story. While heavy fabrics cannot effectively block outdoor street noise due to the mass law, they excel at an entirely different task: absorbing mid-frequency reflections within a room. This breakdown covers the critical divide between transmission loss and noise reduction coefficients, comparing heavy aco...

Why Niche Communities Swear by Weird Mics 21.11.2026

Niche audio communities often fall in love with unusual microphones through happy historical accidents rather than pristine spec sheets. Harmonica players turned the high-impedance Shure Green Bullet into an expressive hand-filtered instrument, while bluegrass groups coordinate live mixes by stepping around vintage-styled Ear Trumpet Labs condensers. Meanwhile, field recordists deploy passive seis...

Why Mediocre Appetizers Make Good Entrees Shine 21.11.2026

Hedonic contrast determines how coursed meals land on the palate. A mediocre opening course can launch an ordinary entree, whereas an outstanding starter sets an impossible standard. Exploring sensory science from Drexel University and a hospitality trial in Hong Kong, this discussion breaks down why narrative influences portion perception, how salt acts as a taste multiplier, and why contrast eff...

Why Ten Seconds Clones Better Than an Hour of Training 20.11.2026

Zero-shot voice cloning uses brief audio as a temporary conditioning signal, while fine-tuning permanently modifies a model's weights. Attempting to improve a clone with an hour of training data often degrades output quality by introducing acoustic noise, style collapse, and catastrophic forgetting. Balancing speaker similarity and clear pronunciation requires techniques like prior-preservation lo...

Choosing Open-Source Frameworks to Fine-Tune TTS 20.11.2026

Fine-tuning text-to-speech models on a single speaker involves navigating legacy architectures and complex licensing. This guide walks through the classic open-source stack, contrasting Tacotron 2's two-stage pipeline with VITS's end-to-end flow. It examines practical realities across maintained forks like Idiap's coqui-tts, Piper's home-automation deployment, XTTS weights restrictions, and StyleT...

How Ribbon and Boundary Microphones Actually Work 20.11.2026

Dynamic and condenser designs dominate audio discussions, but rare microphone categories reveal a deeper map of transducer physics and surface placement. This breakdown examines the corrugated aluminium foil inside ribbon mics, the sixty-second phantom power decay window, and the mechanical figure-eight pattern. It also explores boundary microphones, tracing how flush mounting prevents comb filter...

Why Voice Clones Stay Flat 20.11.2026

Training a professional text-to-speech voice clone requires far more than feeding audio into a model. While zero-shot cloning captures basic timbre, fine-tuning shapes naturalness, emotional range, and behavioral consistency across long passages. Examining data requirements from ElevenLabs, LJ Speech, and CMU ARCTIC shows why recording hours of neutral reading yields a flat clone. Achieving lifeli...

Airing Challengers to Test AI Podcast Writers 20.11.2026

A five-part technical series tested whether rival AI models could replace a baseline scriptwriter when evaluating search primitives. Across fifteen drafts from Qwen 4 Max, Gemini 4 Carbon, Gemini 4 Hydrogen, and Kimi K3.1 Pro, zero scripts failed Claude Opus 5.5's evaluation, though every model ignored topic boundaries and trampled episode fences. By design, challengers aired regardless of baselin...

Between BM25 and Vector Search 20.11.2026

Information retrieval is not just a binary choice between BM25 keyword matching and dense vector embeddings. Across the middle rungs of the retrieval ladder, learned sparse methods like SPLADE expand vocabularies onto inverted indexes while late-interaction systems like ColBERT preserve token-level nuance. A look into how LSA's matrix factorization contrasts with modern neural embedders, and why s...

Reciprocal Rank Fusion and Reranking Eval 20.11.2026

Combining keyword search and vector retrieval requires balancing fundamentally different scales. Reciprocal rank fusion dodges score distortion by using rank positions, though score normalization can sometimes edge it out in relevance benchmarks. Meanwhile, adding an expensive cross-encoder reranker stage does not always guarantee better results and can even degrade retrieval quality depending on...

Why BM25 Still Beats Neural Search on Keywords 20.11.2026

Counting matching words is easy, but ranking them accurately without understanding meaning remains surprisingly resilient. Behind the thirty-year reign of BM25 lies a hand-fitted saturation curve designed to prevent single-word repetition from hijacking queries, paired with inverse document frequency from information theory. Examining document length normalization reveals how search engines mathem...

Exact Search Without a Vector Database 20.11.2026

Building a custom semantic retrieval layer over thousands of podcast transcripts reveals that vector databases are often premature. Designing around speaker turns, preserving attribution, and avoiding topic-averaging failures matter far more than index complexity. The engineering breakdown walks through turnchunk boundaries, local models like Google DeepMind's EmbeddingGemma 2 and Alibaba's Qwen3-...

Why Grep Is Not Enough for Search 20.11.2026

A scan across 22 million words takes less than a tenth of a second with modern tools like ripgrep, yet full-text retrieval still requires an entirely different data structure. Exploring the mechanics of grep, regular expressions, and benchmarks across source code and subtitles reveals where linear scanning hits a wall. The discussion covers why relational B-tree indexes fail on natural text and ho...

Teaching Vision Models to Spot Tiny Architectural Details 20.11.2026

Building an internal tool to look up specific architectural features like bay windows or mansard dormers requires more than just a standard image classifier. Public datasets like FacadeElements, LSAA, and WikiChurches lack fine-grained feature labels, but they provide a strong foundation for negative mining. Using a frozen pretrained backbone like DINOv2 or CLIP with a linear probe beats training...

How Accidental Dataset Formats Became AI Standards 20.11.2026

No standards body designed the dataset formats powering modern AI. They survived because a library shipped a loader or a benchmark needed to score fairly. This breakdown compares speech and image formats across file organization, geometry, and extra flags—from LJSpeech's pipe-delimited CSV and NVIDIA's NeMo JSONL manifests to Mozilla Common Voice, DeepSpeech, COCO, Pascal VOC, and YOLO. Along the...

Why Voice Agents Use Frameworks Like Pipecat 20.11.2026

A 200-millisecond pause can make a caller think the line is dead, yet voice agent builders still add orchestration layers like Pipecat on top of realtime APIs like Gemini Live. Breaking down the four-player market—Pipecat, LiveKit Agents, Vapi, and Retell—reveals what frameworks actually buy you: transport independence over WebRTC, modifying transcripts before an LLM sees them, non-conversation pi...

Bypassing Android's Built-In USB Audio Stack 20.11.2026

Android's built-in USB audio stack is a Linux inheritance topped with a consumer audio server built for ringtones and phone calls. Even with native AudioRecord or AAudio, vendor layers can cap captures at 16 bits, add phantom gain, and apply a hard limiter before an app sees a sample. The episode examines what AOSP actually promises for USB Audio Class 1, why Google's Oboe library relies on a Quir...

Tokyo's Standing Sushi Bars and Market Counters 20.11.2026

Cheap sushi in Tokyo does not mean compromised fish—the gap between budget and luxury is increasingly just seating and ceremony. This guide maps out a full day of eating well without spending a fortune, starting at Toyosu Market before ten and moving to Tokyo's roughly fifty standing sushi bars, or tachigui, including Uogashi Nihon-Ichi, Tachigui Sushi Akira, and Hinatomaru inside Tokyo Station. I...

The Ryukyu Islands Beyond Okinawa's Resort Strip 20.11.2026

The Ryukyu Islands are often sold as a single tropical resort, yet this thousand-kilometer arc holds six endangered languages, two major geographic breaks, and distinct histories from the Ryukyu Kingdom to American administration. A trip through the chain reveals Okinawa's centenarian village of Ogimi, analog whale scouts in the Keramas, Miyako's 152 cultural properties and mud-smearing Paantu fes...

Planning a Murakami Winter in Japan's Onsen 20.11.2026

Planning a winter trip to Japan around atmosphere rather than a checklist starts with six physical ingredients: old wood, a hearth, pressing forest, snow, a valley, and empty streets after dark. While Haruki Murakami's Killing Commendatore house sits near Tokyo, the true northern mood lives in Tohoku's snow-viewing baths. From Ginzan Onsen's new crowd-control rules to Akita's seven-inn Nyuto Onsen...

Why Tokyo Feels Like a Neighborhood of 37 Million 20.11.2026

Tokyo is home to thirty-seven million people, yet it walks like a collection of small neighborhoods. A full-day walking tour through Kuramae, Yanaka, and Shimokitazawa reveals the urbanism behind that scale. From former Edo-period rice granaries turned maker workshops to narrow shotengai shopping streets, Tokyo's permissive zoning and small-lot land ownership protect independent retail, slow down...

Ascultă podcastul My Weird Prompts în Replaio

Radio și podcasturi într-o singură aplicație - gratuit și fără înregistrare. Instaleaz-o azi și nu rata lansarea

Descarcă din Google Play

Replaio nu este editorul podcasturilor; numele emisiunilor, coperțile și materialul audio aparțin autorilor lor și sunt distribuite prin fluxuri RSS publice