SearchSkip to Search Results
- 2Speech perception
- 2audio segmentation
- 2forced alignment
- 2speech recognition
- 1Acoustic analysis
- 7Linguistics, Department of
- 3Linguistics, Department of/Presentations (Linguistics)
- 2Linguistics, Department of/Research Publications (Linguistics)
- 2Temporary Community
- 2Temporary Community/Temporary Collection
- 1Linguistics, Department of/Honours Theses (Linguistics)
The present paper investigates the effect of different inputs on the accuracy of a forced alignment tool built using deep neural networks. Both raw audio samples and Mel-frequency cepstral coefficients were compared as network inputs. A set of experiments were performed using the TIMIT speech...
Poster for the paper "A comparison of input types to a deep neural network-based forced aligner," presented at Interspeech 2018. PAPER ABSTRACT: The present paper investigates the effect of different inputs on the accuracy of a forced alignment tool built using deep neural networks. Both raw...
In recent years, computational modeling has proved to be an essential tool for investigating cognitive processes underlying speech perception (see, e.g., Scharenborg & Boves, 2010). Here we address the question of how an end-to-end computational model that uses the acoustic signal as input...
Natural, spontaneous speech (and even quite careful speech) often shows extreme reduction of many speech segments, even resulting in apparent deletion of consonants. Where the flap ([ɾ]) allophone of /t/ and /d/ is expected in American English, one frequently sees an approximant-like or even...
A number of speech perception studies have been carried out to investigate how we process audio signals containing real words. However, comparatively fewer studies have been conducted looking at how listeners process audio signals containing phonotactically legal pseudowords. Some traditional...
Massive Auditory Lexical Decision: Going Big in the Auditory Domain
Research on silent reading has shown that text genre influences the way texts are read, including differences between prose and poetry (e.g. Zwaan, 1994; Hanauer, 1998). There is little data examining whether text layout (prose vs. poetry) affects the way it is read aloud by non-expert readers,...
A sizable number of phonetic and psycholinguistic experiments have been conducted to investigate the recognition of real words. From this work, researchers have found that various characteristics of lexical items affect the recognition process, such as lexical frequency, phonotactic probability,...
Human production and perception of language, although studied for decades, is largely misunderstood. Furthermore, not all sounds in human language have been studied extensively. Typologically rare sounds arguably lack reliable documentation and research. One such sound is voiceless nasals....