Pronunciation Shadowing Practice

Step 1

Watch and Listen

Watch the short video for context and listen carefully to the target phrase.

Phrase Library

Choose a phrase from the lecture

This activity uses phrases from the lecture "Ben Crystal on Original Pronunciation." Select a phrase from the list below — the target phrase and the original audio in Step 2 will update automatically.

Lecture phrase

Step 2

Study the Original Phrase

Listen to the model audio of the target phrase and study its acoustic shape.

Original waveform

Original amplitude envelope

Original spectrogram (approximate)

Original pitch contour

Step 3

Record Your Shadowing Attempt

Please allow microphone access to record your voice. Recording requires an HTTPS page and a compatible browser.

Idle

My waveform

My amplitude envelope

My spectrogram (approximate)

My pitch contour

Step 4

Compare with the Original

Place the original and your recording side by side and look for differences in length, rhythm, stress, and intonation.

Original phrase — Waveform

Your recording — Waveform

Original phrase — Pitch contour

Your recording — Pitch contour

Overlayed Pitch Contours (DTW alignment)

Blue = Original · Orange = Your recording. The contours are time-normalised and aligned using Dynamic Time Warping (DTW) for direct comparison.

DTW Distance: — Shape correlation: — Direction agreement: — Final tone match: —

Original phrase — Rhythm / amplitude envelope

Your recording — Rhythm / amplitude envelope

The overall similarity score is a pedagogical indicator. It prioritises intonation contour, rhythm and stress. The speech-recognition check verifies whether the target words were pronounced clearly. It is diagnostic and does not replace intonation matching.

Step 5

Read the Feedback

–

Press "Compare Pronunciation" to see your score.

Timing

Record and compare to see feedback on timing.

Rhythm

Record and compare to see feedback on rhythm.

Intonation

Record and compare to see feedback on intonation.

Overall match

Record and compare to see overall feedback.

This automatic feedback is approximate. It is useful for noticing duration, rhythm, stress, and general intonation differences, but it does not replace teacher feedback or professional phonetic analysis.