ClassroomThe classroom
Back to the lesson libraryMECHANISM · 14 MIN
M13.5 CONNECT THE MECHANISM

From a waveform to time-frequency features

Say "buy oat milk" into your phone and it records 16,000 numbers a second. Follow them into the picture of sound that speech models actually read.

LESSON OVERVIEW14 min lesson

Lesson overview

Say "buy oat milk" into your phone and it records 16,000 numbers a second. Follow them into the picture of sound that speech models actually read.

What you’ll explore

  • Waveforms sample amplitude over time; frequency transforms and spectrograms reveal complementary structure, with resolution and sampling assumptions that affect audio learning.
GO TO THE SOURCE

Original explanations, connected to the research.

Speech and Language Processing, 3rd edition draft — chapter on automatic speech recognition (Jurafsky & Martin)Robust Speech Recognition via Large-Scale Weak Supervision (Radford et al., 2022)NumPy documentation — numpy.fft.rfft
Suggest a correction

A precise note can make an explanation better.

Choose the scene and describe what needs attention. Download a feedback file to share through a channel you already use. This page does not send feedback or connect you with a reviewer.

The file includes this note, the scene title, and lesson metadata. Your saved progress and quiz responses are excluded. Download before leaving or reloading to keep your note.