Reverse study
The problem was to identifier the speaker for a given audio
My first approach was the research about state of art of speaker recognition
Then I implemented and tested a few, and choose the best one
The implementation also included the data cleaning and preprocessing part.
After that I need to expose this in a API
Then I read how to build an API in Python and choose flask to it
The api would accept new training data and will score new voices
Shares tags with
- Abstract NodeYou can model knowledge as a graph of nodes of concepts/ideas When you abstract 2 or more ideas in a more general concept finding the edges among those nodes This will make show…
- AtomWe can use atom to describe something that is minimal and indivisible We can have an atom of state like redux recoiljs We can have an atom of thought as Roam and Zettelkästen uses…
- Being wrongYou can be wrong in many situations Make sure to always revalidate your ideas on new information like Bayesian approach P(A I): your probability of something should changed based…
- Bidirectional note takingCreate a note for the source (book) Create a note for each idea/concept of the source For instance, a note for Make it Stick A note for spaced learning A note for interview…
- Bleeding edgeedge Bleeding edge is the state of art of something It is the new technology When you focus on bleeding edge (new), you have an advantage as everybody start in a clean slate. You…
Sign in with GitHub to reply.
Select any passage to quote it.
Comments