Speaker Localization Research Articles

Frontal eye field activity preceding aurally guided saccades.

1. We studied neuronal activity in the monkey's frontal eye field (FEF) in conjunction with saccades directed to auditory targets. 2. All FEF neurons with movement activity preceding saccades to visual targets also were active preceding saccades to auditory targets, even when such saccades were made in the dark. Movement cells generally had comparable bursts for aurally and visually guided saccades; visuomovement cells often had weaker bursts in conjunction with aurally guided saccades. 3. When these cells were tested from different initial fixation directions, movement fields associated with aurally guided saccades, like fields mapped with visual targets, were a function of saccade dimensions, and not the speaker's spatial location. Thus, even though sound location cues are chiefly craniotopic, the crucial factor for a FEF discharge before aurally guided saccades was the location of auditory target relative to the current direction of gaze. 4. Intracortical microstimulation at the sites of these cells evoked constant-vector saccades, and not goal-directed saccades. The direction and size of electrically elicited saccades generally matched the cell's movement field for aurally guided saccades. 5. Thus FEF activity appears to have a role in aurally guided as well as visually guided saccades. Moreover, visual and auditory target representations, although initially obtained in different coordinate systems, appear to converge to a common movement vector representation at the FEF stage of saccadic processing that is appropriate for transmittal to saccade-related burst neurons in the superior colliculus and pons.

Is There Still a Problem About the Self?

The puzzling concept of self is analyzed as a double singularity. Selfi is the self as the bare numerical identity of personhood and self2 is the unique set of beliefs each selfi has about itself. These selves are shown to be discursively produced through the indexical properties of conversation. The dichotomy has been challenged by certain feminist writers. At one level, the challenge can be resolved by treating it as an observation about the multiple selves2 any selfi must handle. A deeper level of challenge suggests that the ways the problems of handling multiple selves2 are dealt with affect the sense of personal identity expressed as selfi. The deeper challenge is met by showing first that the only tenable theory of selfhood is grammatical. The first person pronouns index speech acts with spatial location of speaker and speaker’s position in a moral order. In the light of that demonstration, it is shown how the problems of selves2 can affect the way that pronouns index speech acts with positions in a...

Multiple speaker localization using envelope time delays

It is well known that an array of N sensors can discriminate between at most N-1 continuous sources. However, if the time series produced by the sources contain transients, this limit applies only to samples that are shorter than the duration of the transients. This principle has been applied to the problem of localizing two speakers by using only two microphones. Human speech contains a large number of more or less sharp transients. The time of arrival of the transients at each microphone is estimated by finding the local maxima in the envelope of the time series. The time series at each microphone is replaced with the value of the envelope at the local maxima location, and is set to zero everywhere else. The cross correlation of the two resulting time series presents peaks at the time delays corresponding to the correct location of speakers, whereas the cross correlation of the original time series had a single peak at the wrong time delay. It is shown that the difference is due to the nonlinearity of the operations performed on the time series.

Does the ?Teer? vocalization deter prospecting female red-winged blackbirds?

A speaker-occupation experiment was conducted to determine whether female red-winged blackbirds (Agelaius phoeniceus) use the aggressive “Teer” vocalization to deter conspecific females that are prospecting for opportunities to settle. In each of 14 trials conducted during the breeding seasons of 1987, 1988, and 1989, tape recordings and a pair of loud speakers were used to “defend” suitable habitat. Observations of female red-winged blackbirds settling on these “speaker territories” and on adjacent control areas showed that speaker territories were settled significantly earlier than silent control territories. Each of the 14 first settlers on speaker territories included at least one speaker location within the bird's display area, and 9 of the 14 included both speaker locations. Additional observations showed that the rate at which prospecting females visited control and speaker territories, and the number of females eventually settling on them, were not significantly different. Finally, females that settled on speaker territories sang Teers at significantly higher rates than those on control territories. These results demonstrate that the Teer alone was not sufficient to deter prospecting female red-winged blackbirds. On the contrary, Teers given in the absence of an actual resident may have attracted settlers. It is suggested that the aggressive Teer vocalization might function to establish dominance relationships among the females sharing a polygynous male's territory.

Field playback of male display attracts females in lek breeding sage grouse

Recent correlational studies of lekking sage grouse suggest that male vocal display attracts females. To test this hypothesis further, the natural displays of a territorial male were supplemented with the tape-recorded display of another, reproductively-successful, individual. Significantly, more females approached the speaker's location on days when the recording was played, and also on non-playback days immediately following a playback, than on other non-playback days (Fig. 1). Analysis of male displays indicated that females were responding to the playback itself rather than to changes in male behavior. The “after-response” following a playback suggests that some females present during a playback remembered its location and approached on a subsequent lek visit. The results provide necessary support for the epigamic function of vocal display, and suggest ways in which female responses to male display may influence lek structure.

Simulation of noise distribution in rectangular rooms by means of computer modelling techniques

Models have long been a part of architectural design and have served in both the perceptive and technological development of buildings and spaces. However, physical models can be costly and in some cases difficult to implement. Computer models prove to be powerful, easy to build and both money and time saving. They have been used widely for various energy simulations in architecture, and recently have been used in room acoustics and light distribution in enclosures. This work reviews simulation of sound energy in rooms and discusses the results and difficulties of using this technique. There are two different types of sound energy distribution simulation in rooms; the ray tracing method and the source image method. Both have been implemented successfully, but the latter is more accurate and is easier to use by an architect with little knowledge of acoustics and computers. A computer program is written, using the sound source images method, for rooms of square or rectangular shape. The program can also be extended to simulate rooms with tilted walls, and a sloping ceiling or floor. The program can represent sound field distribution for any number of sound sources or loudspeakers and can calculate sound pressure levels or sound intensity levels at any number of receiving points, at any location inside the enclosure. It also draws the contour lines of the distribution and thus shows the best shape of a room and the best location of speaker and loudspeakers for certain human activities. It also can give the most suitable way of furnishing an enclosure with respect to sound distribution. Other room acoustic terms, such as definition, can also be calculated for that space. The contour lines for these terms can also be drawn.

The Effect of Accent Differences on Comprehension

Several research studies, for example Brown (1968), Wilcox (1978), and Ekong (1982), support the position that the accent which is best understood in ESL/EFL situations is the accent of a good local speaker of English, rather than the accent of a standard native- speaker. Smith and Bisazza's (1982) study is less clear on this point. An experiment conducted in Hong Kong with native Chinese speakers arrived at a different conclusion, namely that the accent of native-speakers (British and American) is comprehended better than that of local Chinese speakers of English. 250 subjects were divided into four equal-proficiency groups and given two tests, a listening comprehension test and a dictation. Each group heard the tests delivered by a different speaker — one American, one British, and two Chinese. On the dictation test especially, the score of the groups who heard the American and British speakers was significantly better than that of the groups which heard the Chinese speakers. Further, the difference was found to be more pronounced with low proficiency subjects than with high proficiency subjects.

Single-unit activity in the auditory cortex of monkeys actively localizing sound sources: Spatial tuning and behavioral dependency

Single-unit recordings were made from the auditory cortex of unanesthetized rhesus monkeys which were trained to perform a sound localization task. Recordings were made from 196 units in 4 animals. Five sound sources were located at 0° (midline) and on either side at azimuths of 37.5° and 75°. Although most units responded to each of the 5 sources, 49% of the units had peak firing rates for the source on the contralateral side at 75° azimuth. Another 26% of the units had peak firing rates at the 37.5° contralateral location, while only 11% had peak rates for the ipsilateral locations. In order to determine whether the behavior of actively locating a sound source affected units in auditory cortex, response rates were compared under two behavioral conditions: one which required the detection of a sound regardless of location and another condition which required identification of a sound's location. Of the 196 units, 16 had different response rates for the two conditions, with 15 of the 16 units having higher evoked rates in the localization task. For the 16 units, the difference in firing rate was typically observed only for a single speaker location. No correlation could be determined between a unit's location within the subdivisions of auditory cortex and its spatial tuning response properties or its dependency upon behavioral condition.

Encoding of sound location by neurons in the auditory cortex of performing rhesus monkeys

Single-unit activity was recorded from the auditory cortex of rhesus monkeys while they localized noise or tone bursts presented from five speakers positioned in a horizontal plane in front of them. One speaker was directly in front of the animal (0°) and two were on each side at azimuths of 37.5° and 75° from the midline. An index of sensitivity to sound location was defined for each unit as the ratio of the neural response rates between the most effective and least effective speaker locations. Out of 196 units, 122 having an index of 2.0 or greater were classified as location-sensitive units. Among the 122 units, 71 (58%) had their highest rate of response for the most contralateral speaker, nine (7%) for the most ipsilateral speaker, and ten (8%) for the center speaker. The most common observation was that response rates gradually decreased as the sound location was shifted from the most contralateral to the most ipsilateral speaker. Only three of the 196 units responded to one speaker location exclusively. [Research supported by NSF.]

Measurement of head movement during auditory localization

A system has been designed to track human‐head movement response to sound originating at different azimuth locations with respect to the head. A videotape record is made of a “point‐source” light carried on a small, lightweight “beanie” mounted on the listener's head. Movement of the head generates an equivalent movement of the light which is being monitored by the video camera and recorded on tape. Also, speaker location, time, and actual sound signal are recorded. Circuitry has been designed which determines the x‐y coordinates of the spot image with respect to the video synch pulses. These are available from a modified video monitor either in real time or from tape replay. This same circuitry interfaces to a computer programmed to take the coordinate and timing information, apply a calibration and process the data into time‐varying head position and velocity. Both digital and graphic outputs are available and examples of graphic printout of typical head movement response are given. System hardware and programs are capable of expansion into 3‐axis operation.

Coding of spatial location by single units in the inferior colliculus of the alert cat.

Effects of altering the spatial location of a tonal sound source were studied in single neurons in the inferior colliculus of unanaesthetized, restrained cats. Speaker position was altered by successive 22.5° displacements in an arc in the horizontal plane around the animal's head. Thresholds for evoked activity in some units were dependent on sound location, but threshold changes for a 22.5° displacement in sound position never exceeded 10 dB. Spike counts of most units changed in a systematic fashion with changes in sound location, the usual pattern of change consisting of a maximum value at some point in the contralateral field and a gradual decline as speaker location was displaced from the most effective position. Discharge patterns of approximately 50% of units were dependent on sound location. The results are compatible with the concept of ‘place coding’ of spatial location.

Auditory localization of value related material.

Neutral words were more accurately located in auditory space than high and low value words by 22 Ss whose values had been established on the Allport, Vernon, and Lindzey test. This effect occurred only at a 45° speaker location; there was no difference between the three types of stimuli at 0°; and the results for non-valued words at other speakers supported these divergent effects. These localization findings partially extend earlier research on visual and auditory thresholds, using an experimental rather than correlational approach. The nature of the underlying perceptual and cognitive processes, however, remains unclear.

The cerci and abdominal giant fibres of the house cricket, Acheta domesticus. I. Anatomy and physiology of normal adults.

The abdominal cerci of crickets are clothed with approximately 3400 hair and campaniform sensilla. Their axons form functional connexions in the terminal abdominal ganglion with a group of ten giant intemeurons. The two largest intemeurons (l. g. i. and m. g. i.) are powerfully excited by ipsilateral sound and air puff stimulation, weakly excited by substrate vibration via non-cercal receptors, and inhibited by contralateral cereal stimulation. These patterns of connexion are used to assay the precision of synapse formation in animals whose cereal development has been experimentally manipulated. The cereal sensilla comprise: about 300 singly innervated filiform hairs and 65 morpho­logically related clavate hairs; 2400 multiply innervated appressed trichoid sensilla of two subtypes; and about 600 campaniform sensilla associated with the sockets of the filiform hairs. The cereal nerve carries a corresponding number of axons with the majority in the 0.1 to 0.5 μ m diameter range. The appressed hairs are rather uniformly distributed over the cereal surface. In contrast, the clavate hairs occur as a single patch on the medial aspect of the base of each cercus. The structure of the socket causes each filiform hair to vibrate in a single plane. The majority vibrate in the longitudinal axis of the cercus, but well defined mid­-dorsal and mid-ventral rows vibrate transversely. Only a few obliquely oriented hairs are found. The axons of the cereal nerve degenerate very rapidly after removal of the cercus, and degeneration figures indicate the distribution of axonal terminals in the ganglion. All terminals in normal animals are ipsilateral, arranged in two loosely organized glomeruli. The l. g. i. and m. g. i. respond to sound stimuli of 2000 Hz and less (best frequency - 600 Hz), to air streams of velocities about 0.5 m/s and above, and to substrate vibration in a narrow band of frequencies centred on about 2000 Hz. Thresholds have not been measured carefully, but are 70 dB or less for sound in an open laboratory, and 10 -6 cm or less for substrate vibration. The l. g. i. and m. g. i. show a marked directional sensitivity to sound, presumably related to the vibration planes of the filiform hairs; differences of 5° in speaker location appear to be discriminable.

The World's Living Languages: Basic Facts of Their Structure, Kinship, Location and Number of Speakers. Siegfried H. Muller

Previous articleNext article No AccessNotes and ReviewsThe World's Living Languages: Basic Facts of Their Structure, Kinship, Location and Number of Speakers. Siegfried H. Muller H. A. Gleason, Jr.H. A. Gleason, Jr. Search for more articles by this author PDFPDF PLUS Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinkedInRedditEmail SectionsMoreDetailsFiguresReferencesCited by International Journal of American Linguistics Volume 31, Number 3Jul., 1965 Article DOIhttps://doi.org/10.1086/464851 Views: 2Total views on this site PDF download Crossref reports no articles citing this article.

A Definition of Proto-Germanic: A Study in the Chronological Delimitation of Languages

It may seem obvious that when we set out to write a grammar of a language, the definition of that language is one of our first tasks. Further, that since we are dealing with language, the criteria used in our definition should be linguistic. These criteria should delimit the language geographically, that is, from other languages, and chronologically, that is, from earlier and later stages of the same language. Yet in spite of these minimum requirements for an adequate definition, our handbooks are not precise in their definitions or explicit in regard to criteria used for the definitions they give. Though we are here concerned with the definition of Proto-Germanic, the problem applies to all languages. Handbooks often delimit a language through the use of social or political changes rather than linguistic changes. Middle High German, for example, has been defined as that stage of German which was spoken between the period of the great medieval German poets and the Reformation. The Norman Conquest and the introduction of printing have been used to define Middle English. While linguistic changes may coincide with other social changes, the coincidences are interesting correlations rather than defining features for the language. We can determine them only after we have selected those differences in language structure which distinguish successive periods of any language, such as Middle English in contrast with Old English and Modern English, or different languages, such as Italian in contrast with French and Sardinian or Low German in contrast with High German and English. Imprecise definitions are objectionable in part because they permit unnecessary disputes to arise regarding the description of a language. Two inadequate definitions from grammars in general use may indicate the need for improved definitions.l Prokosch, CGG 26, contents himself with indicating the location of the Germanic speakers and adding:2 'Shortly before the beginning of our era, the