Audio, Speech and Time Series

1 min
  1. What trade-off does the STFT window size introduce?
  2. Why separate recordings before windowing?
  3. What does WER measure?
  4. What minimal baseline for a seasonal forecast?
  5. [E] Why report the metric by horizon or duration?

Answers. 1. Temporal vs. frequency resolution. 2. To avoid nearly identical windows across partitions. 3. Word errors in transcription. 4. Persistence or seasonal value. 5. An average can mask the failing time zones.

Activity — Anti-leak temporal split

From long recordings, propose a split by person or device before windowing. Draw an example of a leak created by overlapping windows and define a seasonal or persistence baseline. Success: no source segment crosses the partitions.

Preview — the rest of the lesson is for enrolled readers.

Already enrolled with a code?

Your access is tied to your account, not to this link. Sign in with the same email you used in class: your course is waiting, no need to enter the code again.

Sign inNo account yet? Create one
This lesson is part of the “Modern Deep Dives” module

The first modules of the course are open to everyone. For the rest you have three options: buy this course once and for all, subscribe, or enter the code handed out in class.

Are you a student on this course?

The code is tied to your account: sign in or create an account and it will be applied automatically when you come back.

No account yet? Create one