AI and data science

Study Speech Processing and Synthesis with a partner

Speech recognition and text-to-speech, from signals to neural models. Get paired with someone working on Speech Processing for a timed focus session, or drop into the Speech Processing study room.

Topics people work on

  • Speech signals
  • Acoustic modelling
  • Speech recognition
  • Text normalisation
  • Speech synthesis

Getting the most from a Speech Processing session

Run something small

Fit a tiny model or run one experiment in each block. Seeing a result teaches more than another page of theory.

Explain the maths in words

Say what a loss function or an attention head does without equations. If it's fuzzy in words, it's fuzzy in your head.

Keep an error log

Write down each bug or wrong prediction and its cause. Reviewing it in the break builds intuition quickly.

How a focus session works

  1. 1. Choose Speech Processing and Synthesis and a 25 or 50 minute focus length.
  2. 2. We pair you with someone studying Speech Processing, or with someone open to any subject if nobody is online yet.
  3. 3. Say what you'll work on, then focus together by video, voice or text. A short break follows each block.

For adults 18 and over. End the session whenever you like, and block or report anyone in one tap.