background preloader

WaveNet: A Generative Model for Raw Audio

Talking Machines Allowing people to converse with machines is a long-standing dream of human-computer interaction. The ability of computers to understand natural speech has been revolutionised in the last few years by the application of deep neural networks (e.g., Google Voice Search). However, generating speech with computers — a process usually referred to as speech synthesis or text-to-speech (TTS) — is still largely based on so-called concatenative TTS, where a very large database of short speech fragments are recorded from a single speaker and then recombined to form complete utterances. This makes it difficult to modify the voice (for example switching to a different speaker, or altering the emphasis or emotion of their speech) without recording a whole new database. WaveNet changes this paradigm by directly modelling the raw waveform of the audio signal, one sample at a time.

https://deepmind.com/blog/wavenet-generative-model-raw-audio/

Related:  ETLVData Analytics and Science ArticlesDeep learning2

Augmented and Virtual Reality Hub · Research · Tourism, Events and Hospitality Management · Manchester Metropolitan University The Augmented and Virtual Reality Hub investigates new and innovative ways to implement augmented reality in the tourism, museum, art gallery and creative industry context. By collaborating with academic and industry partners, the hub aims to provide a platform for resources as well as seminars in the area of augmented reality and virtual reality. The hub is built on prior expertise gained in the area of e-Tourism. Currently the main focus is on mobile and wearable augmented reality having conducted a number of projects in the museum and art gallery context in Dublin, Cornwall and Manchester.

Apache Spark: A Unified Engine for Big Data Processing By Matei Zaharia, Reynold S. Xin, Patrick Wendell, Tathagata Das, Michael Armbrust, Ankur Dave, Xiangrui Meng, Josh Rosen, Shivaram Venkataraman, Michael J. Franklin, Ali Ghodsi, Joseph Gonzalez, Scott Shenker, Ion Stoica Communications of the ACM, Vol. 59 No. 11, Pages 56-65 10.1145/2934664 Comments The growth of data volumes in industry and research poses tremendous opportunities, as well as tremendous computational challenges.

216 "untranslatable" emotional words from non-English languages University of East London pysch professor Tim Lomas has assembled a list of words referring to emotional states from the world's languages that have no correlate in English. Whether or not that turns out to be true, the list is fascinating. Some favorites below, a surprising number of which are the names of well-known products, projects, or titles: * S'apprivoiser (French) (v.): lit, 'to tame', but a mutual process - both sides slowly learning to trust the other and eventually accepting each other

Vitae 3MT UK semi-finalists 2015 Michael Hall, Life and Health Sciences, Aston University. Localising epileptogenic cortex using magnetoencephalography Chioma Paul, College of Business Arts and Social Sciences, Brunel University. Using fiction to explore the traumatic impact of Grenada's taboo history. FINALIST. See Chioma's presentation at the live final on 8th September.

Ten Ways Your Data Project is Going to Fail Introduction Data science continues to generate excitement and yet real-world results can often disappoint business stakeholders. How can we mitigate risk and ensure results match expectations? Working as a technical data scientist at the interface between R&D and commercial operations has given me an insight into the traps that lie in our path. I present a personal view on the most common failure modes of data science projects. Slides in one pdf here. Pingu, the cute little British/Swiss penguin who... Pingu, the cute little British/Swiss penguin who has been a TV star since 1986, is a great example of how intonation can convey meaning. Nobody in Pingu’s world speaks an identifiable language, but the tone of voice helps convey the story and means that children from many different language backgrounds may enjoy it. Raised voices or more rapid pace may convey excitement or anger.

Visualizing a Job Search or: How to find a Job as a Software Engineer Today, I start a new job at Gusto. This concludes the most comprehensive search I’ve done for a new job. My job search lasted a total of 35 days from start to signed offer letter. Why seek a new job? Talking parrot missing for 4 years found, now speaks Spanish When Nigel vanished four years ago, he spoke with a cultivated British accent. Little is known about where the African grey parrot went, what he did — or who he was with — in those missing years. But when he was reunited with his owner, Darren Chick, in Torrance last week, the British accent was gone and the bird was chattering in Spanish, often mentioning the name “Larry.” The happy reunion began to unfold after veterinarian Teresa Micco — who had been running ads for her own lost African grey parrot, Benjamin — was contacted by the owners of Happy Tails Dog Spa in Torrance, who said they thought they’d found her missing bird at their Torrance home.

Related: