Models of speech synthesis.

The term "speech synthesis" has been used for diverse technical approaches. In this paper, some of the approaches used to generate synthetic speech in a text-to-speech system are reviewed, and some of the basic motivations for choosing one method over another are discussed. It is important to keep in mind, however, that speech synthesis models are needed not just for speech generation but to help us understand how speech is created, or even how articulation can explain language structure. General issues such as the synthesis of different voices, accents, and multiple languages are discussed as special challenges facing the speech synthesis community.

Linguistic aspects of speech synthesis.

The conversion of text to speech is seen as an analysis of the input text to obtain a common underlying linguistic description, followed by a synthesis of the output speech waveform from this fundamental specification. Hence, the comprehensive linguistic structure serving as the substrate for an utterance must be discovered by analysis from the text. The pronunciation of individual words in unrestricted text is determined by morphological analysis or letter-to-sound conversion, followed by specification of the word-level stress contour. In addition, many text character strings, such as titles, numbers, and acronyms, are abbreviations for normal words, which must be derived. To further refine these pronunciations and to discover the prosodic structure of the utterance, word part of speech must be computed, followed by a phrase-level parsing. From this structure the prosodic structure of the utterance can be determined, which is needed in order to specify the durational framework and fundamental frequency contour of the utterance. In discourse contexts, several factors such as the specification of new and old information, contrast, and pronominal reference can be used to further modify the prosodic specification. When the prosodic correlates have been computed and the segmental sequence is assembled, a complete input suitable for speech synthesis has been determined. Lastly, multilingual systems utilizing rule frameworks are mentioned, and future directions are characterized.

Integration of speech with natural language understanding.

The integration of speech recognition with natural language understanding raises issues of how to adapt natural language processing to the characteristics of spoken language; how to cope with errorful recognition output, including the use of natural language information to reduce recognition errors; and how to use information from the speech signal, beyond just the sequence of words, as an aid to understanding. This paper reviews current research addressing these questions in the Spoken Language Program sponsored by the Advanced Research Projects Agency (ARPA). I begin by reviewing some of the ways that spontaneous spoken language differs from standard written language and discuss methods of coping with the difficulties of spontaneous speech. I then look at how systems cope with errors in speech recognition and at attempts to use natural language information to reduce recognition errors. Finally, I discuss how prosodic information in the speech signal might be used to improve understanding.

Processing of speech signals for physical and sensory disabilities.

Assistive technology involving voice communication is used primarily by people who are deaf, hard of hearing, or who have speech and/or language disabilities. It is also used to a lesser extent by people with visual or motor disabilities. A very wide range of devices has been developed for people with hearing loss. These devices can be categorized not only by the modality of stimulation [i.e., auditory, visual, tactile, or direct electrical stimulation of the auditory nerve (auditory-neural)] but also in terms of the degree of speech processing that is used. At least four such categories can be distinguished: assistive devices (a) that are not designed specifically for speech, (b) that take the average characteristics of speech into account, (c) that process articulatory or phonetic characteristics of speech, and (d) that embody some degree of automatic speech recognition. Assistive devices for people with speech and/or language disabilities typically involve some form of speech synthesis or symbol generation for severe forms of language disability. Speech synthesis is also used in text-to-speech systems for sightless persons. Other applications of assistive technology involving voice communication include voice control of wheelchairs and other devices for people with mobility disabilities.

Commercial applications of speech interface technology: an industry at the threshold.

Speech interface technology, which includes automatic speech recognition, synthetic speech, and natural language processing, is beginning to have a significant impact on business and personal computer use. Today, powerful and inexpensive microprocessors and improved algorithms are driving commercial applications in computer command, consumer, data entry, speech-to-text, telephone, and voice verification. Robust speaker-independent recognition systems for command and navigation in personal computers are now available; telephone-based transaction and database inquiry systems using both speech synthesis and recognition are coming into use. Large-vocabulary speech interface systems for document creation and read-aloud proofing are expanding beyond niche markets. Today's applications represent a small preview of a rich future for speech interface technology that will eventually replace keyboards with microphones and loud-speakers to give easy accessibility to increasingly intelligent machines.

Growth and viability of macrophages continuously stimulated to produce nitric oxide

Deregulated production of nitric oxide (NO) has been implicated in the development of certain human diseases, including cancer. We sought to assess the damaging potential of NO produced under long-term conditions through the development of a suitable model cell culture system. In this study, we report that when murine macrophage-like RAW264.7 cells were exposed continuously to bacterial lipopolysaccharide (LPS) or mouse recombinant interferon-γ (IFN-γ) over periods of 21–23 days, they continued to grow, but with doubling times 2 to 4 times, respectively, longer than the doubling time of unstimulated cells. Stimulated cells produced NO at rates of 30 to 70 nmol per million cells per day throughout the stimulation period. Within 24 hr after removal of stimulant, cells resumed exponential growth. Simultaneous exposure to LPS and IFN-γ resulted in decreased cell number, which persisted for 2 days after removal of the stimulants. Exponential growth was attained only after an additional 4 days. Addition of N-methyl-l-arginine (NMA), an NO synthase inhibitor, to the medium inhibited NO production by 90% of all stimulated cells, partially reduced doubling time of cells stimulated with LPS or IFN-γ, and partially increased viability and growth rates in those exposed to both LPS and IFN-γ. However, when incubated with LPS and IFN-γ at low densities both in the presence and in the absence of NMA, cells grew at a rate slower than that of unstimulated cells, with no cell death, and they resumed exponential growth 24 hr after removal of stimulants. Results from cell density experiments suggest that macrophages are protected from intracellularly generated NO; much of the NO damaging activity occurred outside of the producer cells. Collectively, results presented in this study suggest that the type of cellular toxicity observed in macrophages is markedly influenced by rate of exposure to NO: at low rates of exposure, cells exhibit slower growth; at higher rates, cells begin to die; at even higher rates, cells undergo growth arrest or die. The ability of RAW264.7 cells to produce NO over many cell generations makes the cell line a useful system for the study of other aspects of cellular damage, including genotoxicity, resulting from exposure to NO under long-term conditions.

A Cell-free System to Study Regulation of Focal Adhesions and of the Connected Actin Cytoskeleton

Assembly and modulation of focal adhesions during dynamic adhesive processes are poorly understood. We describe here the use of ventral plasma membranes from adherent fibroblasts to explore mechanisms regulating integrin distribution and function in a system that preserves the integration of these receptors into the plasma membrane. We find that partial disruption of the cellular organization responsible for the maintenance of organized adhesive sites allows modulation of integrin distribution by divalent cations. High Ca2+ concentrations induce quasi-reversible diffusion of β1 integrins out of focal adhesions, whereas low Ca2+ concentrations induce irreversible recruitment of β1 receptors along extracellular matrix fibrils, as shown by immunofluorescence and electron microscopy. Both effects are independent from the presence of actin stress fibers in this system. Experiments with cells expressing truncated β1 receptors show that the cytoplasmic portion of β1 is required for low Ca2+-induced recruitment of the receptors to matrix fibrils. Analysis with function-modulating antibodies indicates that divalent cation-mediated receptor distribution within the membrane correlates with changes in the functional state of the receptors. Moreover, reconstitution experiments show that purified α-actinin colocalizes and redistributes with β1 receptors on ventral plasma membranes depleted of actin, implicating binding of α-actinin to the receptors. Finally, we found that recruitment of exogenous actin is specifically restricted to focal adhesions under conditions in which new actin polymerization is inhibited. Our data show that the described system can be exploited to investigate the mechanisms of integrin function in an experimental setup that permits receptor redistribution. The possibility to uncouple, under cell-free conditions, events involved in focal adhesion and actin cytoskeleton assembly should facilitate the comprehension of the underlying molecular mechanisms.

Approaches to enhancing the retroviral transduction of human synoviocytes

This report concerns a clinical trial for rheumatoid arthritis (RA), approved by the US National Institutes of Health and the Food and Drug Administration. An amphotropic retrovirus (MFG-IRAP) was used ex vivo to transfer a cDNA encoding human interleukin-1 receptor antagonist (IL-1Ra) to synovium. The protocol required the transduced cells to secrete at least 30 ng IL-1Ra/106 cells per 48 h before reimplantation. Here we have evaluated various protocols for their efficiency in transducing cultures of human rheumatoid synoviocytes. The most reliably efficient methods used high titer retrovirus (approximately 108 infectious particles/ml). Transduction efficiency was increased further by exposing the cells to virus under flow-through conditions. The use of dioctadecylamidoglycylspermine (DOGS) as a polycation instead of Polybrene (hexadimethrine bromide) provided an additional small increment in efficiency. Under normal conditions of static transduction, standard titer, clinical grade retrovirus (approximately 5 × 105 infectious particles/ml) failed to achieve the expression levels required by the clinical trial. However, the shortfall could be remedied by increasing the time of transduction under static conditions, transducing under flow-through conditions, or transducing during centrifugation.

An economic analysis of unilateral refusals to license intellectual property

The intellectual property laws in the United States provide the owners of intellectual property with discretion to license the right to use that property or to make or sell products that embody the intellectual property. However, the antitrust laws constrain the use of property, including intellectual property, by a firm with market power and may place limitations on the licensing of intellectual property. This paper focuses on one aspect of antitrust law, the so-called “essential facilities doctrine,” which may impose a duty upon firms controlling an “essential facility” to make that facility available to their rivals. In the intellectual property context, an obligation to make property available is equivalent to a requirement for compulsory licensing. Compulsory licensing may embrace the requirement that the owner of software permit access to the underlying code so that others can develop compatible application programs. Compulsory licensing may undermine incentives for research and development by reducing the value of an innovation to the inventor. This paper shows that compulsory licensing also may reduce economic efficiency in the short run by facilitating the entry of inefficient producers and by promoting licensing arrangements that result in higher prices.

The evolution of RNA viruses: A population genetics view

RNA viruses are excellent experimental models for studying evolution under the theoretical framework of population genetics. For a proper justification of this thesis we have introduced some properties of RNA viruses that are relevant for studying evolution. On the other hand, population genetics is a reductionistic theory of evolution. It does not consider or make simplistic assumptions on the transformation laws within and between genotypic and phenotypic spaces. However, such laws are minimized in the case of RNA viruses because the phenotypic space maps onto the genotypic space in a much more linear way than on higher DNA-based organisms. Under experimental conditions, we have tested the role of deleterious and beneficial mutations in the degree of adaptation of vesicular stomatitis virus (VSV), a nonsegmented virus of negative strand. We also have studied how effective population size, initial genetic variability in populations, and environmental heterogeneity shapes the impact of mutations in the evolution of vesicular stomatitis virus. Finally, in an integrative attempt, we discuss pros and cons of the quasispecies theory compared with classic population genetics models for haploid organisms to explain the evolution of RNA viruses.

Differential Control of Xanthophylls and Light-Induced Stress Proteins, as Opposed to Light-Harvesting Chlorophyll a/b Proteins, during Photosynthetic Acclimation of Barley Leaves to Light Irradiance

Barley (Hordeum vulgare L.) plants were grown at different photon flux densities ranging from 100 to 1800 μmol m−2 s−1 in air and/or in atmospheres with reduced levels of O2 and CO2. Low O2 and CO2 partial pressures allowed plants to grow under high photosystem II (PSII) excitation pressure, estimated in vivo by chlorophyll fluorescence measurements, at moderate photon flux densities. The xanthophyll-cycle pigments, the early light-inducible proteins, and their mRNA accumulated with increasing PSII excitation pressure irrespective of the way high excitation pressure was obtained (high-light irradiance or decreased CO2 and O2 availability). These findings indicate that the reduction state of electron transport chain components could be involved in light sensing for the regulation of nuclear-encoded chloroplast gene expression. In contrast, no correlation was found between the reduction state of PSII and various indicators of the PSII light-harvesting system, such as the chlorophyll a-to-b ratio, the abundance of the major pigment-protein complex of PSII (LHCII), the mRNA level of LHCII, the light-saturation curve of O2 evolution, and the induced chlorophyll-fluorescence rise. We conclude that the chlorophyll antenna size of PSII is not governed by the redox state of PSII in higher plants and, consequently, regulation of early light-inducible protein synthesis is different from that of LHCII.

Reverse two-hybrid and one-hybrid systems to detect dissociation of protein-protein and DNA-protein interactions.

Macromolecular interactions define many biological phenomena. Although genetic methods are available to identify novel protein-protein and DNA-protein interactions, no genetic system has thus far been described to identify molecules or mutations that dissociate known interactions. Herein, we describe genetic systems that detect such events in the yeast Saccharomyces cerevisiae. We have engineered yeast strains in which the interaction of two proteins expressed in the context of the two-hybrid system or the interaction between a DNA-binding protein and its binding site in the context of the one-hybrid system is deleterious to growth. Under these conditions, dissociation of the interaction provides a selective growth advantage, thereby facilitating detection. These methods referred to as the "reverse two-hybrid system" and "reverse one-hybrid system" facilitate the study of the structure-function relationships and regulation of protein-protein and DNA-protein interactions. They should also facilitate the selection of dissociator molecules that could be used as therapeutic agents.

Transformation of Plasmodium falciparum malaria parasites by homologous integration of plasmids that confer resistance to pyrimethamine.

Plasmodium falciparum malaria parasites were transformed with plasmids containing P. falciparum or Toxoplasma gondii dihydrofolate reductase-thymidylate synthase (dhfr-ts) coding sequences that confer resistance to pyrimethamine. Under pyrimethamine pressure, transformed parasites were obtained that maintained the transfected plasmids as unrearranged episomes for several weeks. These parasite populations were replaced after 2 to 3 months by parasites that had incorporated the transfected DNA into nuclear chromosomes. Depending upon the particular construct used for transformation, homologous integration was detected in the P. falciparum dhfr-ts locus (chromosome 4) or in hrp3 and hrp2 sequences that were used in the plasmid constructs as gene control regions (chromosomes 13 and 8, respectively). Transformation by homologous integration sets the stage for targeted gene alterations and knock-outs that will advance understanding of P. falciparum.

Microevolutionary responses in experimental populations of plants to CO2-enriched environments: parallel results from two model systems.

Despite the critical role that terrestrial vegetation plays in the Earth's carbon cycle, very little is known about the potential evolutionary responses of plants to anthropogenically induced increases in concentrations of atmospheric CO2. We present experimental evidence that rising CO2 concentration may have a direct impact on the genetic composition and diversity of plant populations but is unlikely to result in selection favoring genotypes that exhibit increased productivity in a CO2-enriched atmosphere. Experimental populations of an annual plant (Abutilon theophrasti, velvetleaf) and a temperate forest tree (Betula alleghaniensis, yellow birch) displayed responses to increased CO2 that were both strongly density-dependent and genotype-specific. In competitive stands, a higher concentration of CO2 resulted in pronounced shifts in genetic composition, even though overall CO2-induced productivity enhancements were small. For the annual species, quantitative estimates of response to selection under competition were 3 times higher at the elevated CO2 level. However, genotypes that displayed the highest growth responses to CO2 when grown in the absence of competition did not have the highest fitness in competitive stands. We suggest that increased CO2 intensified interplant competition and that selection favored genotypes with a greater ability to compete for resources other than CO2. Thus, while increased CO2 may enhance rates of selection in populations of competing plants, it is unlikely to result in the evolution of increased CO2 responsiveness or to operate as an important feedback in the global carbon cycle. However, the increased intensity of selection and drift driven by rising CO2 levels may have an impact on the genetic diversity in plant populations.