With conventional earphone listening, accurate simulation of the direction of sound arrival is impossible because the sound field is not fixed in space but rotates with the head. A system for overcoming this limitation has been developed. Signals are transmitted by electromagnetic or underwater radiation to two directional receptors attached to the listener's head. The receptors are connected to the earphones through a Ster-Bin network. This network divides each of the signals from the receptors into two parts. One part is fed into its corresponding earphone; the other is delayed with respect to the first by the time interval required for sound to travel normally between the two ears, and is fed into the opposite earphone. In addition, both parts are modified at high frequencies to simulate the diffraction effects attributable to the head. The varying relative signal strengths from the directional receptors, as a function of the direction of signal arrival, modified by the Ster-Bin network, provide the listener with orientation capability. Further apparatus improvement and applications to aerospace and underwater swimmers' communications devices are envisioned.
When driven from a purely reactive source, conventional noise figure is undefined for a two-port network such as an audio amplifier. In order to assess the noise performance of small-signal amplifiers driven from reactive sources such as capacitor microphones, a signal-to-wideband-average-noise ratio is defined. It is shown that for transducers where output is related to internal reactance by a simple transformer relationship and where the Thevenin equivalent internal impedance is a pure series and or , there exists an optimum value of internal reactance for maximum signal-to-noise ratio when used with a given noisy amplifier. The equations for optimum reactance and signal-to-average-noise ratio are general and include low-frequency dependence of the transistor noise generators, equalization or frequency dependence of amplifier gain, and source resistance as well as reactance. Solutions are given for the practical audio case with typical low-noise audio transistors for both inductive and capacitive transducers.
Recent experimental work at Telefunken-Decca in the correction of tracing error using harmonic generation is seen to find its justification in the well-known power-series expansions of that error. An exact expansion, in which the structure of the general term is evident, is newly derived in an apt form for mechanization. A process diagram is given.
Pitch perturbations are defined as normally minute variations in the lengths of adjacent pitch periods of voiced speech. It has been shown that perturbations increase in size and number in the presence of laryngeal abnormalities such as cancerous growths. Previous studies have been limited by the unavailability of electronic equipment of sufficient accuracy to perform the pitch measurements in real time. The apparatus to be described represents an intermediate step in the design of a real-time perturbation detector. Operating in an expanded time scale and using regenerated pulse trains derived from visual measurement of voiced waveforms, perturbation measurements are made which are comparable with the computer-calculated values.
Dissipation of output devices in an ideal Class B amplifier can range from 10 to 400 percent of maximum available sine-wave output power. Dissipation is calculated as a function of output for a variety of signals and loads under steady-state and instantaneous conditions.
Though "typical" patterns can be abstracted from speech sounds, the actual patterns vary widely. The human mind extracts many patterns, making its final selection on the basis of meaning. Until computers are capable of semantic analysis, voice-operated equipment must rely on "constrained" speech, which restricts itself to a simply abstracted pattern. A vowel meter has been built which abstracts such a pattern from vowel sounds. With this equipment and others of a similar nature, a training program will be undertaken to determine the effects of training upon speech precision, in the hope that such training will make possible a voice-operated brailler for communication with the deaf-blind. The underlying principles of the vowel meter are explained.
Manned space flight uses one of the oldest communications media, speech, and a voice recorder is required to store it. CBS Laboratories developed the voice-time recorders for the NASA Gemini missions, and flights have proved their effectiveness. The small, light, reliable device records the mixed voices of the astronauts on one track and a digital time signal on the other. Cartridges with one hour capacity can be changed during orbiting, which permits taking as many cartridges as the mission duration requires. Tape is 0.110 inch wide, moves at 0.6 in/s and is contained in a unique coaxial-reel bidirectional cartridge. An ac synchronous motor, energized from a dc to ac converter, drives the mechanism at constant speed. Ground players/duplicators allow convenient transcription and time correlation of the cartridges.
A FET operated so that channel pinch-off is avoided behaves like a voltage controlled variable resistor. This paper considers a "voltage divider" attenuator using the device in this way. Attenuation (in decibels) varies almost linearly with control voltage over a range. Distortion may be less than 0.01 percent, temperature effects are usually negligible, and the network can operate satisfactorily up to a few megahertz. Circuits of this kind demand a FET having special characteristics.
Communication channel capacity is almost invariably at a premium between space vehicles and the earth. In the Apollo moon exploration program, plans call for the use of a special audio processing technique to enhance the signal-to-noise ratio. Extremely deep space probes such as the Mars Mariner IV now use very low bit rate transmissions because of signal-to-noise considerations: future manned planetary missions will benefit greatly from effective speech bandwidth compression techniques. This paper,calls attention to the interest in speech processing in future space exploration programs.
This paper covers the design of magnetic tape wear simulators for high-resolution (wide-band) tape and general instrumentation type tape. Validation was performed by comparing tape wear on each simulator to tape wear on tape transports of each type (wide-band and general purpose). Tape wear on the simulators as well as on transports was judged by the degradation in the dropout characteristics. Recommendations were presented for specification tape wear requirements based upon simulator use.
The characteristics of Bessel, Butterworth, and Chebyshev filters are briefly reviewed. The "Rauch" RC active low-pass filter configuration is described and design formulae are obtained. Tables of normalized capacitor values are provided for Bessel, Butterworth, and ±½ dB, ±1 dB, ±2 dB, and ±3 dB Chebyshev filters. Two examples indicate the use of the tables.
An electronic delay modulator that operates by scanning a fixed multitap delay line has been devised, and an experimental model has been built and tested. The device is ultimately capable of delay-variation-times-signal-bandwidth products as high as 11, with minimal amplitude modulation and good modulation linearity. The frequency at which delay can be modulated is limited only by the speed of the scanning circuit and not by the reciprocal of the mean delay, as in conventional delay modulators which vary the propagation time of an entire line. The behavior of an experimental device of this type is described and applications are suggested.
The importance of basic planning for sound amplification and reinforcement systems is discussed. The first step in basic planning is to determine the functions of the sound system by a study of the building and use requirements for the system. A list of objectives or "functional requirements" will serve as the basis for sound system planning and can be interpreted by signal flow diagrams. These can then serve as the basis for the more detailed functional diagrams or wiring diagrams of the sound system. After determining the functional requirements of the system, the designer proceeds with the acoustical and architectural planning. A basic decision is the choice of either a central or a distributed loudspeaker system. This is followed by a determination of the location and orientation of microphones and loudspeakers and the selection of the actual transducers required for the job. Finally, control and amplification equipment are selected. Basic planning is presented as a portion of the overall effort that includes the detailed electronic design of the system, a specification if one is required, careful installation, and final adjustments.
