The video and slides from Simon’s keynote are now online under Courses > One-off events.
November 1, 2022
Bitrate
The bitrate (or bit rate) of a signal is the number of bits required to store, or transmit, 1 s of that signal. A bit is a binary number: either 0 or 1. Let’s calculate the bitrate of a digital waveform. First you should revise the concepts of sampling and quantisation from this module of the […]
Continue reading... October 11, 2014
Windowing
When we say that a signal is non-stationary we mean that its properties, such as the spectrum, change over time. To analyse signals like this, we need to first assume that these properties do not change over some short period of time, called the frame. We can then analyse individual frames of the signal, one at a […]
Continue reading... November 15, 2014
Token passing
Token passing is a really nice way to understand (and even to implement) Viterbi search for Hidden Markov Models. Here we see token passing in action, and you can look at the spreadsheet to see the calculations. To keep things simple, we are ignoring transition probabilities in this example. It would be simple to add them […]
Continue reading... October 11, 2014
Spectrum and spectrogram
The spectrum and the spectrogram are much more useful ways of analysing speech signals than the waveform. We look at how to create them using Wavesurfer and what effect the analysis window size has on what we see.
Continue reading... October 18, 2014
TD-PSOLA …the hard way
Time-Domain Pitch Synchronous Overlap and Add (TD-PSOLA) can modify the fundamental frequency and duration of speech signals, without affecting the segment identity – that is, without changing the formants. Normally, it’s an automatic algorithm, but here we do it the hard way – by hand! If you want to follow-along, you will need Audacity and these materials (a […]
Continue reading... January 2, 2016
Interactive unit selection
Just a toy demo, but should give you some idea of how unit selection waveform generation works. Click with your mouse to choose a candidate diphone from each column, then the corresponding synthesised waveform will appear. You can click on the synthesised waveform to hear it again. Try to obtain the most natural-sounding synthesis by […]
Continue reading... October 11, 2014
Classification and regression trees (CART)
A quick introduction to a very simple but widely-applicable model that can perform classification (predicting a discrete label) or regression (predicting a continuous value). The tree is learned from labelled data, using supervised learning. Before watching this video, you might want to check that you understand what Entropy is.
Continue reading... February 1, 2015
Autocorrelation for estimating F0
Most methods for estimating F0 start from autocorrelation. The idea is pretty simple: we are just looking for a repeating pattern in the waveform, which corresponds to the periodic vocal fold activity. For some waveforms, it might be possible to do that directly in the time domain, but in general that doesn’t work very well. […]
Continue reading... October 30, 2015
Wave propagation on the surface of water
At the Alhambra (Granada, Spain) I saw this nice example of waves from a point source propagating in all directions at a fixed speed.
Continue reading... October 11, 2014
Sampling and quantisation
Is digital better than analogue? Here we discover that there are limitations when storing waveforms digitally. We learn that the consequence of sampling at a fixed rate is an upper limit on the frequencies that can be represented, called the Nyquist frequency. In addition to the limitations of sampling, storing each sample of the waveform as a […]
Continue reading... October 11, 2014
Pipeline architecture for TTS
Most text-to-speech systems split the problem into two main stages. The first stage is called the front end and contains many separate processes which gradually build up a linguistic specification from the input text. The second stage typically uses language-independent techniques (although they still require a language-specific speech corpus) to generate a waveform. Here we see those two […]
Continue reading...
This is the new version. Still under construction.