Speech recognition of foreign accented speech is one of the most difficult tasks in ASR. The problem of foreign accent is addressed in this study using acoustic models of the target language phonemes (French phonemes in our case) adapted with speech data from 3 other languages: English (US and UK), German and Spanish. Recognition results obtained for 11 language groups of speakers show that error rate can be significantly reduced when standard acoustic models of phonemes are adapted using speech data from other languages. Phonological rules are also introduced into the standard phonetic description of the lexical units to account for some foreign accent pronunciation variants. It appears that using phonological rules together with foreign language adapted acoustic units provides the best recognition performance. The highest error rate reduction (40%) is obtained on English speakers. 1.

This dissertation investigates articulatory setting (AS), a language's underlying or default posture of the articulators (i.e., the tongue, jaw, and lips). Inter-speech posture (ISP) of the articulators (the position of the articulators when they are motionless during inter-utterance pauses) is used as a measure of AS in Canadian English and Québécois French. The dissertation reports two experiments using a combination of Optotrak and ultrasound imaging to test whether ISP is language specific in both monolingual and bilingual speakers, whether it is affected by phonetic context, and whether it is influenced by speech mode (monolingual or bilingual). Results of Experiment 1 show significant differences in ISP across the English and French monolingual groups, with English exhibiting a higher tongue tip, more protruded upper and lower lips, and narrower horizontal lip aperture. Results also show that for English speakers, the jaw ISP is somewhat influenced by phonetic context while the lip and tongue ISP are not. For French speakers, only certain lip components of ISP are influenced by phonetic context while the ISP of the tongue and jaw are not.

The present study investigated the perception and production of English /w / and /v / by native speakers of Sinhala, German, and Dutch, with the aim of examining how their native language phonetic processing affected the acquisition of these phonemes. Subjects performed a battery of tests that assessed their identification accuracy for natural recordings, their degree of spoken accent, their relative use of place and manner cues, the assimilation of these phonemes into native-language categories, and their perceptual maps (i.e., multidimensional scaling solutions) for these phonemes. Most Sinhala speakers had near-chance identification accuracy, Germans ranged from chance to 100 % correct, and Dutch speakers had uniformly high accuracy. The results suggest that these learning differences were caused more by perceptual interference than by category assimilation; Sinhala and German speakers both have a single native-language phoneme that is similar to English /w / and /v/, but the auditory sensitivities of Sinhala speakers make it harder for them to discern the acoustic cues that are critical to /w/-/v / categorization.

This paper briefly reviews state of the art related to the topic of speech variability sources in automatic speech recognition systems. It focuses on some variations within the speech signal that make the ASR task difficult. The variations detailed in the paper are intrinsic to the speech and affect the different levels of the ASR processing chain. For different sources of speech variation, the paper summarizes the current knowledge and highlights specific feature extraction or modeling weaknesses and current trends. 1.

Research on the acquisition of the ability to pronounce a second language has been dominated by acoustic studies as opposed to articulatory studies (e.g., Birdsong, in press; Flege et al., 2003;

Abstract- This paper proposes an acoustic phonetic study of the foreign accents in the Arabic language. To analyze on a large scale of the connected variations, the contribution of the automatic tools acoustico-phonetic decoding tools along the adaptation forces. Based on the results of the uses of these tools, we managed to differentiate between three major forms of accents: The English language speaker accent (American and British population’s accent), the European people accent (French speaker accent, German speaker accent) and the Persian speaker accent.

learn to differentiate /r–l / on the basis of F3 onset frequency? ∗

Abstract not found