text-to-speech

Christopher Hitchens delivers a speech defending freedom of speech, including the right of audiences to hear offensive speakers and a take on the fire-in-a-crowded-theatre line. It is framed around the Charlie Hebdo attack.

Raph Levien presents Xi editor at RustConf 2016, a high-performance text editor built in Rust. The talk goes into technical details, building polished GUI apps, and running a community-focused open source project.

A Key & Peele sketch in which two friends keep misreading the tone of each other's text messages while making plans.

Russ Cox gives a 23-minute tour of the Acme text editor from Plan 9, showing its mouse-driven interface, executing commands from text, and its filesystem-based interface to other programs.

A HandmadeCon 2016 panel where Chris Hecker, John Miles, Sean Barrett and Chris Green give an oral history of software texture mapping in 90s games.

Casey Muratori captures CygnusEd Professional running on an Amiga 2000, showing how smoothly it scrolled text. He notes the real effect is better on a period CRT.

A demonstration of Larynx, an offline open source neural text-to-speech system supporting nine languages. The video itself is narrated with a Larynx-generated voice.

A Burnistoun sketch from BBC Scotland in which two men try to use a voice-recognition elevator that cannot understand their Scottish accents.

Microsoft Research's Rick Rashid demonstrates a speech recognition and translation system, built on deep neural networks, that turns his spoken English into Chinese speech in 2012.

The closing speech from Charlie Chaplin's 1940 film The Great Dictator, in which the barber appeals for humanity, kindness and liberty over machinery and cleverness.

A restored recording of Steve Jobs unveiling the original Macintosh in 1984, including the moment it is taken out of its bag and speaks. The upload comes from the team that rediscovered and restored the lost footage.

A short status-update demo from Figure showing its humanoid robot using OpenAI-powered speech-to-speech reasoning to converse and perform tasks.

A Burnistoun sketch in which a voice-recognition elevator cannot understand two Scottish men saying the number eleven. It satirizes voice interfaces that fail on accents.

Andrej Karpathy builds the GPT tokenizer from scratch, implementing byte pair encoding along with encode and decode. He shows how many odd LLM behaviors trace back to tokenization, and covers Unicode byte encodings along the way.

Rob Miles explains "glitch" tokens, odd strings in large language model vocabularies that make the model produce gibberish. He covers how they likely arise from tokenizer training data, such as Reddit usernames, that never appeared in the model's own training.

The first lecture of Stanford's CS231n Winter 2016 course on convolutional neural networks for visual recognition, introducing the course and the historical context of computer vision.

FortNine looks at why Tesla Autopilot has rear-ended motorcyclists, speculating that the vision system misreads motorcycle taillights as a distant car. It argues that radar and other sensors would help.

Short Airbus footage of its ATTOL project, in which a test aircraft performed a fully automatic, vision-based take-off at Toulouse-Blagnac. The crew ran eight take-offs over several hours.

Michael Reeves rigs up a camera and software that tracks a goldfish's movements to choose stock trades, tied to wallstreetbets sentiment. It is a comedic programming and electronics stunt.

The first lecture of N J Wildberger's History of Mathematics course, starting with Pythagoras' theorem and following Stillwell's text. It emphasizes the flow of ideas over proofs.

Archival footage of Larry Tesler demonstrating Gypsy, the modeless text editor he and Tim Mott built at Xerox PARC, running on the Computer History Museum's restored Alto. It is a long hands-on walkthrough of one of the earliest graphical document editors.

A TVO current affairs panel debates Canada's Bill C-16, which added gender identity and expression to human rights law and the Criminal Code. Jordan Peterson argues it threatens free speech, and other guests respond.

James Powell's nearly two-hour PyData talk builds up Python's data model, decorators, generators and context managers, coding live in Vim. It aims to take intermediate programmers toward real Python expertise.

Jason Scott discusses making his documentary GET LAMP at a Google Tech Talk and screens part of the film. It covers the history of text adventures and interactive fiction, with interviews of Infocom developers and others.

Ted Nelson, the originator of hypertext and Project Xanadu, gives a personal account at age 80 of his ideals, his career, and how a poor career plan led to discoveries and adventures. It works as an alternative history of personal computing and hypertext.

Jonathan Blow gives a formal, hour-long demo of his Jai compiler focused on the implicit context feature. The context carries things like allocators through a program without explicit parameter passing, and the talk walks through the code.

Tom Gally, a former professional translator and University of Tokyo professor, discusses in April 2023 how GPT-4 and similar models may affect translators' careers. He draws on Japanese-English translation experience.

The SIGGRAPH 2005 demo of dual photography from Stanford, which swaps the roles of projector and camera to render a scene from the light's point of view. It includes the reveal of a hidden playing card.

A player piano is made to reproduce human speech, here reciting a European Environmental Criminal Court proclamation in German at the 2009 World Venice Forum. This upload adds subtitles so the words become understandable.

A 2007 video of someone trying to write Perl scripts using Windows Vista speech recognition, with comically bad results. The footage shows the recognizer repeatedly mangling commands.

Deviant Ollam's two-hour conference talk explains how elevators work, from the pit and buffers to the car top, controller logic and access-control features. It shows how this knowledge can be used for optimization and for bypassing building security.

Climate Town argues for moving away from gas stoves. It covers the gas industry's decades of marketing and lobbying, health effects of gas cooking, methane leaks, and the case for induction.

Tom Scott investigates the claim that the microwave was invented to heat hamsters, and interviews James Lovelock about freezing and microwave-thawing hamsters in the 1950s. It traces early magnetron uses beyond radar.

Steve Blank's Computer History Museum lecture argues that Silicon Valley's roots lie in WWII radar and electronic warfare work, Stanford's Fred Terman and Cold War military funding rather than the semiconductor alone.

A Computer History Museum oral history with Avie Tevanian covering his math talent, work on the Mach microkernel at CMU, joining NeXT, leading NEXTSTEP development and the 1996 Apple acquisition.

Kevlin Henney's GOTO 2018 talk argues that much of what looks new in software is a recycling of concepts with long histories. It draws on history to show the slow, steady current beneath constant churn in frameworks and trends.

Tom Scott tells how John Reber's plan to dam San Francisco Bay was disproved by the Army Corps of Engineers' giant hydraulic Bay Model in Sausalito. The model showed the plan would have been dangerously wrong.

Weber State's Professor Kelly explains the Toyota RAV4 Hybrid transaxle, including its planetary gearset eCVT and all nine operating modes. It also covers how the engine reaches nearly 40 percent thermal efficiency.

Christof Paar's university lecture introducing Galois fields and finite-field arithmetic as used in AES. It covers the mathematics behind the AES S-box.

Rowan Atkinson's speech at the launch of the Reform Section 5 campaign in Parliament in 2012. He argues against criminalizing insulting words and for answering speech with more speech.