JPH06186980A - Sounding body driving circuit - Google Patents

Abstract

PURPOSE:To improve the quality of an output speech by dispersing the frequency spectrum of noise components included in an output speech signal obtained from a sounding body. CONSTITUTION:An energy converting circuit 10 inputs three-bit digital speech data and converts the data into a signal having 1 or >=2 pulses dispersed within sampling time width corresponding to the digital speech data, i.e., a signal having those pulses at different time base positions, and is equipped with a PWM pulse output terminal 10a and a sign bit output terminal 10b. Namely, the rise timing of the pulses outputted from the energy converting circuit 10 is dispersed in the time-base direction. The frequency spectrum of noises is dispersed, so noise components outside the audio frequency range increase and noises within the audio frequency range relatively decrease to the contrary. At this time, high frequency noise components are smoothed by the impedance (capacity and resistance components) of the sounding body 7.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a piezoelectric element (PIEZ).
O) as a sounding body to generate a sound.

[0002]

2. Description of the Related Art When an audio signal voltage having a waveform as shown in FIG. 5 is sampled in a time width t, it becomes a group of voltages of various levels as shown in FIG. 6 arranged in the time axis direction. The energy within each unit time t at this time corresponds to the area of the shaded portion. Then, as shown in FIG. 7, this energy can be converted into a pulse width (constant level) within the same unit time width t.

Therefore, when a sounding body is driven by digitized sound data, conventionally, the digital sound data is converted into a pulse width-modulated PWM pulse as shown in FIG. 7, and this PWM pulse is directly generated. Was being applied to.

FIG. 8 is a diagram showing the drive circuit.
Is a pulse width modulation circuit for converting 3-bit digital audio data into 1-bit PWM pulse. The remaining 1 bit of the input data is the polarity (positive,
Negative) is a sign bit. 1a is a PWM pulse output terminal and 1b is a sine bit output terminal. Reference numerals 2 and 3 are NAND gates, 4 to 6 are inverters, and 7 is a sounding body made of a piezoelectric element.

In this pulse width conversion circuit 1, as shown in FIG. 9, when the input digital audio data is "0" (absent) in accordance with the pulse of the conversion fundamental frequency shown in (a), (b). Although the output pulse is zero as shown in FIG. 5, when the digital audio data is “1”, the PWM pulse having a pulse width of 0.5 cycle of the fundamental frequency pulse is output as shown in (c). When the digital audio data is "2", a PWM pulse having a pulse width for one cycle of the fundamental frequency pulse is output as shown in (d). Further, when the digital audio data is "3", a PWM pulse having a pulse width of 1.5 cycles of the fundamental frequency pulse is output as shown in (e). Further, when the digital audio data is "4", a PWM pulse having a pulse width of two cycles of the fundamental frequency pulse is output as shown in (f). Similarly, when the digital audio data is "7", a PWM pulse having a pulse width of 3.5 cycles of the basic frequency pulse is output as shown in (i). Then, assuming that the time for 3.5 cycles is the sampling time width t described above, an audio energy signal of 8 gradations can be obtained from the pulse width modulation circuit 1.

In the PWM pulse thus obtained from the pulse width modulation circuit 1, the output side of the inverter 5 is "H" and the output side of the inverter 6 is "H" when the sign bit is "H" only during the period of the pulse width. When it becomes “L”, the sounding body 7 is driven. When the sign bit is "L", the sounding body 7 is driven with the opposite polarity.

[0007]

However, in this conventional circuit, the rising edges of the PWM pulses obtained by the pulse width modulation circuit 1 are all at the same timing. For this reason,
There is a problem that the sampling frequency determined by the sampling time width t, harmonic noise that is an integral multiple of the sampling frequency, repetitive noise, and the like are mixed into the voice signal as fixed noise, and the quality of the output voice obtained by the sounding body 7 deteriorates.

It is an object of the present invention to provide a sounding body drive circuit which is not affected by such a sampling frequency and has improved output voice quality.

[0009]

To this end, the present invention provides a sounding body driving circuit for converting a digital audio data into an energy signal distinguished by an occupied time width within a specific sampling time width and driving a sounding body. The energy signal is a signal in which one or more pulses are present at different time axis positions within the sampling time width according to the digital audio data.

[0010]

EXAMPLES Examples of the present invention will be described below. FIG. 1 is a circuit diagram of a sounding body driving circuit of the embodiment. In this embodiment, an energy conversion circuit 10 is used instead of the pulse width modulation circuit 1 of FIG. The energy conversion circuit 10 inputs 3-bit digital audio data, and a signal in which one or more pulses are scattered in correspondence with the digital audio data within a sampling time width t, that is, those pulses are time-based. The signal is converted into an ergi signal which is a signal that is made to exist at different positions. 10a
Is a PWM pulse output terminal, and 10b is a sign bit output terminal.

The operation of the energy conversion circuit is performed as follows. As shown in FIG. 2, the input digital audio data is “0” according to the conversion fundamental frequency shown in (a).
When (none), the output pulse is zero as in the conventional case as shown in (b), but when the digital audio data is "1", as shown in (c), 0.5 cycle of the fundamental frequency pulse. The pulse having the pulse width of is output in the intermediate time zone of the sampling time width t. When the digital audio data is "2", two 0.5-cycle pulses of the fundamental frequency pulse are output at symmetrical time positions around the intermediate time of the sampling time width t as shown in (d). To do. Hereinafter, similarly, a pulse having a waveform symmetrical with respect to the intermediate time of the sampling time width t is output according to the digital audio data.

As a result, in this embodiment, the rising timings of the pulses output from the energy conversion circuit 10 are dispersed in the time axis direction. Therefore, since the frequency spectrum of noise is dispersed, the noise component outside the audible frequency range increases, and conversely, the noise within the audible frequency range relatively decreases.

At this time, the high-frequency noise component is smoothed by the impedance (capacitance or resistance component) of the sounding body, and can be cut by using a simple low-pass filter if necessary. Therefore, the quality of the voice obtained from the sounding body 7 is improved.

FIG. 3 shows the frequency spectrum of the signal obtained when data (triangular waveform of frequency 1 kHz) is input to the circuit of this embodiment (FIG. 1), and FIG. 4 is the same as the circuit of the conventional example (FIG. 8). It is a figure which shows the frequency spectrum of the signal obtained when the data of is input. In both cases, the sampling frequency is 6.25 KHz (t = 0.16 ms), and the conversion frequency is 200 KHz (width of one pulse = 5 μs). The signal levels are all -7 dB (1 KHz), but the noise level is -21 dB (5.1 KHz) in the example of FIG. 3 of the present embodiment, whereas it is -6 dB in the conventional example of FIG. It is as large as (6.25 KHz). As described above, the noise component of the embodiment of FIG. 3 has a lower level and is more scattered.

In the above embodiment, one pulse is generated within the sampling time width t, for example, when the data is "1", but the pulse width is made narrower for similar data "1". It is also possible to generate a plurality of pulses dispersed at equal intervals within the time t. When the conversion frequency is 200 kHz and one pulse (pulse width 5 μs), four pulses (pulse width 1.25 μs) can be generated by quadrupling the frequency to 800 KHz. In this way, if the conversion frequency is increased, the noise component is dispersed over a wider range, and the adverse effect of noise on the voice band is further reduced.

[0016]

As described above, according to the present invention, since the frequency spectrum of the noise component included in the output voice signal obtained from the sounding body is dispersed, the quality of the output voice can be greatly improved. There are great advantages.

[Brief description of drawings]

FIG. 1 is a block diagram of a sounding body driving circuit according to an embodiment of the present invention.

FIG. 2 is an operation explanatory diagram of an energy conversion circuit of the drive circuit.

FIG. 3 is a diagram of a frequency spectrum of a signal obtained when 1 KHz triangular waveform data is input to the circuit of the present embodiment.

FIG. 4 is a diagram of a frequency spectrum of a signal obtained when 1 KHz triangular waveform data is input to the conventional circuit.

FIG. 5 is a sound signal waveform diagram.

FIG. 6 is an explanatory diagram of audio signal sampling.

FIG. 7 is an explanatory diagram of sampling an audio signal using pulse width modulation.

FIG. 8 is a block diagram of a conventional sounding body driving circuit.

FIG. 9 is an operation explanatory diagram of the pulse width modulation circuit of the drive circuit.

[Explanation of symbols]

1: pulse width modulation circuit, 2, 3: NAND gate, 4 to
6: Inverter, 7: Sound generator (PIEZO), 10: Energy conversion circuit.

Claims (1)

[Claims] 1. A sounding body drive circuit for driving a sounding body by converting digital sound data into an energy signal distinguished by an occupied time width within a specific sampling time width, wherein the energy signal is converted into the digital sound data. Accordingly, the sounding body drive circuit is characterized in that one or two or more pulses are present in the sampling time width with different time axis positions.