Linguistics 478/578
Speech Technology

Handouts:

• Syllabus
• Survey
• Assignment #1
• IPA handout
• Acoustics handout
• Assignment #2
• Assignment #3
• Assignment #4
• Assignment #5
• Anonymized results for assignment #4
• Assignment #6
• Assignment #7
• revised Baum-Welch handout
• Final Assignment

Matlab files:

• Corrected simplespectrogram.m: simple spectrogram file. (Incorrect original file)
• snd2spec.m: spectrum from recording
• spec.m: spectrum from wave
• mhfdatoolexamplefda: fdatool example
• myfile.m: m-file demo
• quantdemo.m: quantization demo
• sampledemo.m: sample rate demo
• lpcdemo: LPC demo
• aeC.wav: consonant cues
• getpitch.m: m-file to calculate pitch
• ioCa.wav: locality of nasalization
• lpcpitch.m: LPC pitch modification
• mha.wav: steady-state [a]
• psolapics.m: PSOLA demo (pics)
• psolapitchdemo.m: PSOLA demo (pitch)
• torment.wav: pitch and stress
• tV.wav: vowel cues
• dowords.m: framework for replacement rules (for HW#4)
• ed.m: calculates edit distance (for HW#4)
• digits.zip: digits data file
• trn.m: encoding framework
• tst.m: testing framework
• wlengthtrn.m: length encoding
• wlengthtst: length testing
• fbdemo.m: filterbank demo
• filterbank.m: filterbank encoding
• fbdiff.m: filterbank testing
• stand.wav: example for durational changes
• stand.TextGrid: textgrid for the above
• dtwdemo.m: graphical DTW demo
• dtw.m: required to run DTW demo
• dtwvec.m: DTW implementation for encoding framework
• logspecdemo.m: Scaling demo for front-end analysis
• mhi.wav: me saying [i]
• melfbdemo.m: mel scale filterbank demo
• cepdemo.m: graphical cepstrum demo
• lpcrep.m: LPC implementation for encoding framework
• nounsverbs.m: for assignment #6
• nouns.txt: for assignment #6
• verbs.txt: for assignment #6
• kmeansdemo.m: simple demo for k-means clustering
• allvecs.m: vector quantization for a digits dataset encoding (revised: 4/2/13)
• allvecsFS.m: vector quantization for a digits dataset with a final state symbol appended
• hmmtst.m: HMM-based cross-validation test for digits dataset
• phonenumbers.txt: dataset for final assignment (random numbers form UA linguistics website)
• firstlast.m: continuous distribution demo
• names.txt: file required for preceding demo (random names from UA linguistics website)

Links:

• Praat (free)
• Matlab (glitzy and expensive, but free for UA folks)
• IPA font for MSword users and TIPA IPA font for LaTeX users (The tipa fonts are part of the core TEX distribution, but you can add them on with that link if your distribution doesn't include them.)
• On-line IPA keyboard if you use MSWord and have the patience for this!
• VTAR matlab package to calculate and visualize vocal tract filtering.
• Classic older synthesis examples
• More recent synthesis examples
• Foma finite-state toolkit
• The Monty Hall problem