BEGIN:VCALENDAR VERSION:2.0 PRODID:-//128.220.36.25//NONSGML kigkonsult.se iCalcreator 2.26.9// CALSCALE:GREGORIAN METHOD:PUBLISH X-FROM-URL:https://www.clsp.jhu.edu X-WR-TIMEZONE:America/New_York BEGIN:VTIMEZONE TZID:America/New_York X-LIC-LOCATION:America/New_York BEGIN:STANDARD DTSTART:20231105T020000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RDATE:20241103T020000 TZNAME:EST END:STANDARD BEGIN:DAYLIGHT DTSTART:20240310T020000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RDATE:20250309T020000 TZNAME:EDT END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT UID:ai1ec-21275@www.clsp.jhu.edu DTSTAMP:20240328T111235Z CATEGORIES;LANGUAGE=en-US:Student Seminars CONTACT: DESCRIPTION:Abstract\n\n\n\nAutomatic discovery of phone or word-like units is one of the core objectives in zero-resource speech processing. Recent attempts employ contrastive predictive coding (CPC)\, where the model lear ns representations by predicting the next frame given past context. Howeve r\, CPC only looks at the audio signal’s structure at the frame level. The speech structure exists beyond frame-level\, i.e.\, at phone level or eve n higher. We propose a segmental contrastive predictive coding (SCPC) fram ework to learn from the signal structure at both the frame and phone level s.\n\nSCPC is a hierarchical model with three stages trained in an end-to- end manner. In the first stage\, the model predicts future feature frames and extracts frame-level representation from the raw waveform. In the seco nd stage\, a differentiable boundary detector finds variable-length segmen ts. In the last stage\, the model predicts future segments to learn segmen t representations. Experiments show that our model outperforms existing ph one and word segmentation methods on TIMIT and Buckeye datasets. DTSTART;TZID=America/New_York:20220211T120000 DTEND;TZID=America/New_York:20220211T131500 LOCATION:Ames Hall 234 @ 3400 N. Charles Street\, Baltimore\, MD 21218 SEQUENCE:0 SUMMARY:Student Seminar – Saurabhchand Bhati “Segmental Contrastive Predict ive Coding for Unsupervised Acoustic Segmentation” URL:https://www.clsp.jhu.edu/events/student-seminar-saurabhchand-bhati/ X-COST-TYPE:free X-ALT-DESC;FMTTYPE=text/html:\\n\\n
\\nAbstr act
\n\n\n\n\nAutomatic discovery of phone or word-like units is one of the core objectives in zero-resource speech processing. Recent attempts employ contrastive predictive coding (CPC)\, where the model learns repre sentations by predicting the next frame given past context. However\, CPC only looks at the audio signal’s structure at the frame level. The speech structure exists beyond frame-level\, i.e.\, at phone level or even higher . We propose a segmental contrastive predictive coding (SCPC) framework to learn from the signal structure at both the frame and phone levels.\n\n\nSCPC is a hierarchical mode l with three stages trained in an end-to-end manner. In the first stage\, the model predicts future feature frames and extracts frame-level represen tation from the raw waveform. In the second stage\, a differentiable bound ary detector finds variable-length segments. In the last stage\, the model predicts future segments to learn segment representations. Experiments sh ow that our model outperforms existing phone and word segmentation methods on TIMIT and Buckeye datasets.
Abstr act
\nZipf’s law is commonly glossed by the aphorism “infre quent words are frequent\,” but in practice\, it has often meant that ther e are three types of words: frequent\, infrequent\, and out-of-vocabulary (OOV). Speech recognition solved the problem of frequent words in 1970 (wi th dynamic time warping). Hidden Markov models worked well for moderately infrequent words\, but the problem of OOV words was not solved until sequ ence-to-sequence neural nets de-reified the concept of a word. Many other social phenomena follow power-law distributions. The number of native sp eakers of the N’th most spoken language\, for example\, is 1.44 billion ov er N to the 1.09. In languages with sufficient data\, we have shown that monolingual pre-training outperforms multilingual pre-training. In less-f requent languages\, multilingual knowledge transfer can significantly redu ce phone error rates. In languages with no training data\, unsupervised A SR methods can be proven to converge\, as long as the eigenvalues of the l anguage model are sufficiently well separated to be measurable. Other syst ems of social categorization may follow similar power-law distributions. Disability\, for example\, can cause speech patterns that were never seen in the training database\, but not all disabilities need do so. The inabi lity of speech technology to work for people with even common disabilities is probably caused by a lack of data\, and can probably be solved by find ing better modes of interaction between technology researchers and the com munities served by technology.
\nBiography
\nMark Hasegawa-Johnson is a William L. Everitt Faculty Fellow of Electrical and Computer Engineering at the University of Illinois in Urbana-Champaig n. He has published research in speech production and perception\, source separation\, voice conversion\, and low-resource automatic speech recogni tion.
\n X-TAGS;LANGUAGE=en-US:2022\,December\,Hasegawa-Johnson END:VEVENT END:VCALENDAR