Topics for foundation lecture on 1st November 2018

[Simon]

Foundation lecture: maths & probability

If there are things you would like to cover, please add them to this topic.

[Fabian R]

I know that probably is out of scope of this course, but thinking in dissertation or in the aim of having a deeper understanding when we read some paper I think it would be way useful to work in matrix transformations and understand it properly (how matrix rotation works, how to translate data etc…) and concepts as eigenvectors, eigenvalues, eigendecompositions (thinking about PCA, jFA, i-vectors and so on).

Maybe if this is too out of scope it can be interesting to have the mathematical tools to understand Expectation Maximization in depth. For that aim, I think that for a student without math background can be interesting to see how derivatives works and how to relate them in an optimization problem, see some optimization problem for a convex function or something like that and relate them with the maximum likelihood problem for speech. Is this proper for the course? 🙂

[Simon]

For Speech Processing, we don’t really need much Linear Algebra beyond vectors. There will be some more advanced material in Speech Synthesis, where we will use Linear Algebra operations (affine transforms) to adapt Gaussians to new data.

[Simon]

Deriving optimisation algorithms (e.g., Expectation Maximisation) is also beyond the scope of the course. But, I’d be willing to offer an additional session on this if there is enough interest – please survey your fellow students and let me know how many would like this.

[Daniel]

I’ll second this – but as someone new to the elements described above I worry it may be too much for a single hour. It might also be good to do an Information-Theoretic account of entropy if that is something that you think will be useful as well?

[Simon]

Information Theory is a powerful tool, but well beyond the scope of Speech Processing. I’d be happy to help you one-on-one or in a small group, if this is something you are trying to understand.

[Author]

Posts