Looking at Spectrogram in Praat cs 4706 Jan

Praat • http: //www. fon. hum. uva. nl/praat/ • Developed by Paul Boersma and

Praat • Pros: designed for speech analysis (not only sound edition or spectrogram visualization),

Let’s… • • See the pitch contour of Solfege: Compare the spectrograms of 1.

Task 1 • Record a file with all the vowels and look at the

Task 2 • Record files with each consonant class as defined by manner of

Task 3 • Record something in a very loud voice, to produce clipping, and

Task 4 • Create a sound from equation with noise, and try to filter

Task 5 • record speech with some noise (hum, transients) in the background and

Task 6 • Record a sample of the same short sentence as angry speech,

Slides: 16

Download presentation

Looking at Spectrogram in Praat cs 4706, Jan 30 Fadi Biadsy

Praat • http: //www. fon. hum. uva. nl/praat/ • Developed by Paul Boersma and David Weenink Institute of Phonetic Sciences University of Amsterdam The Netherlands • General purpose speech tool: editing, segmentation and labeling, audio DSP, prosodic manipulation

Praat • Pros: designed for speech analysis (not only sound edition or spectrogram visualization), nice GUI, scripting, active development and community, prosodic manipulation • Cons: limited scripting language, slow GUI

Let’s… • • See the pitch contour of Solfege: Compare the spectrograms of 1. It’s hard to recognize speech 2. It’s hard to wreck a nice beach • • Record something with clipping Record “it’s cool to play with praat” – – Manipulate the pitch contour Annotate this utterance

Task 1 • Record a file with all the vowels and look at the formants to see how they differ

Task 2 • Record files with each consonant class as defined by manner of articular/voiced and unvoiced, all in the context of the same vowel, to see what the spectrograms look like: /pax/, /tax/, /kax/; /bax/, /dax/, /gax/

Task 3 • Record something in a very loud voice, to produce clipping, and see what the waveform looks like

Task 4 • Create a sound from equation with noise, and try to filter the noise out

Task 5 • record speech with some noise (hum, transients) in the background and see was the spectrogram tells them; try filter out the noise.

Task 6 • Record a sample of the same short sentence as angry speech, sad speech, happy speech, and see what things (pitch contour, pitch mean and max, intensity mean and max, spectral information) differ. choose something fully voiced if possible.