Use Of Speech Synthesis Research Articles

The symmetric technique of formant transition generation for use in speech synthesis in Arabic

The symmetric technique of formant transition generation for use in speech synthesis in Arabic

The BiLSTM-based synthesized speech recognition

Homer Dudley's VODER is considered the first attempt to synthesize human speech electronically by breaking it down into acoustic components. Fifty years later, Terminator 2 featured an example of human speech synthesized with artificial intelligence that was used to deceive a human. Speech synthesis is the artificial simulation of human speech using a computer or other device. The counterpart of the voice recognition, speech synthesis is mainly used to convert textual information into audio information so that a person can naturally interact with digital devices. For example, it is used in assistive technology to help visually impaired people read textual content. A separate direction is the use of speech synthesis to create a clone of a person's voice. Deepfake voice technology, also called voice cloning, has advanced to the point where it can accurately reproduce the human voice by mimicking intonation and other features of the speaker. And it can be used to harm a person. Attackers can employ it to fool voice authentication systems or create fake audio recordings to defame public figures, or combine voice clone with social engineering techniques to bamboozle people. This article discusses the architecture of a voice recognition system that will significantly reduce the possibility of fraud using deepfake voice.

Naming practice app on tablets and communication robots for people with aphasia

Aphasia is a language disorder that negatively affects a person's ability to understand and generate language. Although it has been said that patients in the chronic phase are not likely to recover their ability, a number of recent studies report that highly frequent and continuous training enables recovery. Therefore, we have started a project that aims to improve the environment in which people with aphasia can continue language learning at home. In this presentation, we demonstrate naming practice apps, using speech recognition technology for people with aphasia. They can check whether their utterances are accurate, through the use of speech recognition and they can listen to the correct pronunciation of words through the use of speech synthesis. Field trials using the app for tablets have been conducted since Dec. 2014, and for the communication robot, Pepper, since Dec. 2015. Although the average of speech recognition accuracy was 61 % (false acceptance: 2%, false rejection: 37%), the participants accepted the poor robustness of the speech recognition, and tried to speak clearly. Some participants felt as if the robot was somewhat similar to a child, and others understood its inflexibility, and so in consequence they both accepted its low speech recognition ability.

Razvoj zbirke slovenskega čustvenega govora iz radijskih iger – EmoLUKS

V prispevku predstavljamo graditev slovenske zbirke čustvenega govora za umetno tvorjenje govora in hkrati raziščemo tudi možnosti njene uporabe pri razpoznavanju čustvenega stanja govorca. V prispevku se osredotočamo na opis razvite metodologije za označevanje paralingvistične informacije v govoru na primeru označevanja čustvenih stanj v slovenskih radijskih igrah. Zbirka vsebuje govorne zvočne signale sedemnajstih radijskih iger. Trenutno označeno gradivo obsega čustven govor enega govorca in ene govorke. Čustvene oznake posnetkov smo pridobili s pomočjo dvostopenjskega označevanja s petimi prostovoljnimi označevalci, ki so označili posnetke v dveh časovno ločenih intervalih. Način označevanja omogoča medsebojno primerjavo oznak označevalcev. S pomočjo označenega gradiva v obeh iteracijah poročamo o konsistentnosti označevalcev in ujemanju njihovih mnenj. Na podlagi večinskega mnenja pridobljenih čustvenih oznak vsakemu posnetku pripišemo tisto čustveno oznako, ki je bila med označevalci največkrat izbrana, in tako označene posnetke združimo v zbirko čustvenega govora EmoLUKS, ki jo kvantitativno in kvalitativno ovrednotimo z uporabo uveljavljenega samodejnega sistema za razpoznavanje čustvenih stanj govorca. Konsistentnost oznak ovrednotimo z dvorazrednim in sedemrazrednim od govorca odvisnim razvrščevalnikom čustvenih stanj. Uspešni rezultati razpoznavanja dodatno potrjujejo, da podatkovna zbirka kljub svoji zahtevnosti vsebuje jasno izražena čustvena stanja govorca.

Improving Literacy Skills and Differentiating Learning Speed among Primary School Children through a Computer-Assisted Learning Tool

Our paper relates the development of an innovative speech synthesis tool to improve literacy skills in CP- and CE1-level primary school pupils in France (i.e. children aged 6-8 years in the first two years of primary education). This tool is a software already available on a Web 2.0 platform. Its main objective is to deepen understanding of the link between phonological decoding and encoding in the production of written texts, by taking into account correct/incorrect written responses, which is made possible by the use of speech synthesis as a means of real-time reviewing and self-correction of written work by audio feedback. In our paper, we present some key concepts associated with speech synthesis and the rationale behind the proposed exercises. Also, to assess whether the activities have a positive impact on writing, the experimental phases have been conducted in 7 classes using the platform. Video data have been collected in each class in order to identify how the platform is used as well as difficulties encountered by teachers and pupils. Interviews with the users have helped us to determine the advantages and the obstacles of this computer-assisted learning method. Based on these study results, we developed additional functions to improve the tool, that we will present at the end of our paper.

Method and system for training a text-to-speech synthesis system using a domain-specific speech database

A system, method and computer readable medium that trains a text-to-speech synthesis system for use in speech synthesis is disclosed. The method may include recording audio files of one or more live voices speaking language used in a specific domain, the audio files being recorded using various prosodies, storing the recorded audio files in a speech database; and training a text-to-speech synthesis system using the speech database, wherein the text-to-speech synthesis system selects audio selects audio segments having a prosody based on at least one dialog state and one speech act.

Non-linear frequency scale mapping for voice conversion in text-to-speech system with cepstral description

Voice conversion, i.e. modification of a speech signal to sound as if spoken by a different speaker, finds its use in speech synthesis with a new voice without necessity of a new database. This paper introduces two new simple non-linear methods of frequency scale mapping for transformation of voice characteristics between male and female or childish. The frequency scale mapping methods were developed primarily for use in the Czech and Slovak text-to-speech (TTS) system designed for the blind and based on the Pocket PC device platform. It uses cepstral description of the diphone speech inventory of the male speaker using the source-filter speech model or the harmonic speech model. Three new diphone speech inventories corresponding to female, childish and young male voices are created from the original male speech inventory. Listening tests are used for evaluation of voice transformation and quality of synthetic speech.

System and method for personalizing electronic mail messages by rendering the messages in the voice of a predetermined speaker

Electronic messages are personalized by rendering the messages to recipients in the voice of a predetermined human speaker. This is accomplished by associating with the message payload a set of basis vectors comprising speech parameters from the predetermined human speaker for use in speech synthesis.

Effects of discourse context on the intelligibility of synthesized speech for young adult and older adult listeners: applications for AAC.

The use of speech synthesis in electronic communication aids allows individuals who use augmentative and alternative communication (AAC) devices to communicate with a variety of partners. However, communication will only be effective if the speech signal is readily understood by the listener. The intelligibility of synthesized speech is influenced by a variety of factors, including the provision of context. Although the facilitative effects of context have been demonstrated extensively in studies with young adults, there are few investigations into older adults' ability to decode the synthesized speech signal. The present study investigated whether discourse context affected the intelligibility of synthesized sentences for young adult and older adult listeners. Listeners were asked to repeat 15-word sentences that were either presented in isolation or preceded by a story that set the context for the sentence. Participants correctly repeated significantly more words in the sentences when they were preceded by related sentences than when the sentences were presented in isolation. This research shows a facilitating effect of context in discourse, wherein previous words and sentences are related to later sentences, for both younger and older adult listeners. These results have direct implications for AAC system message transmission.

Heterogeneous relation graphs as a formalism for representing linguistic information

Heterogeneous relation graphs (HRGs) can be used to represent arbitrary linguistic information. Originally designed for use in speech synthesis, HRGs can be used for speech annotation purposes also. In the HRG formalism, atomic linguistic entities such as words, syllables and phones are represented by attribute value matrices known as linguistic items. Using attribute value matrices for items allows them to contain any type or amount of linguistic information. Items are organised into linguistic relations, which take the form of lists, trees or other structures. Items can belong to more than one relation, which allows words to appear in the word relation and, say, the syntax relation. The ability to have items in multiple relations, along with a function ability which can calculate certain values on the fly, eliminates much of the redundancy present in simpler systems. The HRG formalism is not tied to any particular linguistic theory, nor does it impose any preset ideas about what sort of format syntax, prosody or phonology information should have. This paper explains the HRG formalism in detail, and shows why we think this is superior to the types of “multi-level” formats normally used in speech synthesis and database annotation.

Temporal organization of segmental and prosodic features of speech in the word accent of common Japanese

The acoustic–phonetic characteristics of the word accent in common Japanese are based on the moraic structure by which segmental and prosodic features are temporally coordinated. The present study investigates the temporal organization of these features with an emphasis on the influence of ‘‘moraic phonemes.’’ The speech material consists of words of varying lengths, phonetic constituents, and accent types embedded in a carrier sentence. The timing of the prosodic features was represented by the onset/offset of the accent command, estimated from the F0 contour using a quantitative model for the process of F0 contour generation. The timing of the segmental features was represented by the acoustic–phonetic boundaries, detected by referring to the frequency–time–intensity patterns. While the timing of the accent command is strongly correlated with the onset of the vowel rather than with the onset of the mora-initial consonant, segmental constituency, especially the presence/absence of the moraic phonemes, is found to exert strong systematic influences on their relative timing. The results are formulated as a quantitative model of timing for use in speech synthesis.

The use of speech synthesis in exploring different speaking styles

The possibility to vary speaker type and speaking style will be a feature of the next generation text-to-speech systems. Already today, the need for these possibilities is apparent in dialogue systems and when speech synthesis is used as prostheses for persons with a communication handicap. Much of the information needed is not yet available. In this contribution we argue that speech synthesis itself is an efficient tool to study and understand the variability in speech. Some different methods are reviewed, representing both analysis/synthesis techniques and signal manipulations as well as text-to-speech. The main emphasis is, on the work at KTH, to study the variation of speaker and speaking styles in the context of our text-to-speech system.

Animating speech: an automated approach using speech synthesised by rules

This paper is concerned with the problem of animating computer drawn images of speaking human characters, and particularly with the problem of reducing the cost of adequate lip synchronisation. Since the method is based upon the use of speech synthesis by rules, extended to manipulate facial parameters, and there is also a need to gather generalised data about facial expressions associated with speech, these problems are touched upon as well. Useful parallels can be drawn between the problems of speech synthesis and those of facial expression synthesis. The paper outlines the background to the work, as well as the problems and some approaches to solution, and goes on to describe work in progress in the authors' laboratories that has resulted in one apparently successful approach to low-cost animated speaking faces. Outstanding problems are noted, the chief ones being the difficulty of selecting and controlling appropriate facial expression categories: the lack or naturalness of the synthetic speech; and the need to consider the body movements and speech of all characters in an animated sequence during the animation process.

Human Factors and Synthetic Speech

Recent advances in linguistics, speech science, psychology, and especially in computers have made unlimited text-to-speech conversion systems a practical reality. However, the use of audio output from a computer poses special problems in ergonomics, most of which have not been dealt with in the literature. To provide high quality speech synthesis in a useful and usable system requires research in several areas (prototype applications, general design issues for touch-tone phone/speech interfaces, and improving the speech signal itself). In this paper, we review recent work in our laboratory in each of these areas. We close with a discussion of the implications of our work to date for the development and use of speech synthesis.

Syntezator mowy jako narzędzie wspierające pracę tłumacza

The speech synthesiser as a tool supporting the job of the translator
 The present paper discusses the speech synthesis technology and its potential use as a tool supporting translators. The growing popularity, improving quality and ubiquity of speech synthesis on the one hand and continuously falling prices of solutions that adopt this technology lead to a situation in which the use of synthesizers as a tool supporting the job of translators becomes a real perspective. The article suggest the use of speech synthesis for proof reading and editing of translated documents. Speech synthesis can help eliminate those mistakes that cannot be detected by auto correction, such as some spelling errors, punctuation mistakes, mistakes in proper names, lacking words. The article also presents advantages and disadvantages of using synthetic speech.