GB2310984A

GB2310984A - Digital audio processing

Info

Publication number: GB2310984A
Application number: GB9609782A
Authority: GB
Inventors: Paul Anthony Frindle; Peter Charles Eastty
Original assignee: Sony United Kingdom Ltd
Current assignee: Sony Europe Ltd
Priority date: 1996-03-08
Filing date: 1996-05-10
Publication date: 1997-09-10
Anticipated expiration: 2016-05-10
Also published as: GB2310983A; GB2310983B; GB2310985A; GB2310985B; GB9609783D0; GB2310984B; GB9609774D0; GB9605004D0; GB9609782D0

Description

1 DIGITAL AUDIO PROCESSING 2310984 This invention relates to the

processing of digital audio signals.

In many modem audio signal processing devices, such as audio mixing consoles, audio processing operations which had previously been carried out in the analogue domain are now performed on sampled digital audio signals.

An example of this is the "dynamics" section of a digital audio mixing console.

Dynamics processing refers to a family of processing techniques generally having a non-linear effect on the audio signal (compared with the substantially linear techniques of simple gain adjustment and additive mixing). In general, effects classified as "dynamics" tend to have a distorting effect on the sound represented by the audio signal, albeit often a pleasing or useful distortion. For example, the gain applied to an audio signal might be non-linearly adjusted so that the audio signal has a substantially constant level - alleviating the lcvel variations which might result as a performer moves towards and away from a microphone.

Here, the transition from analogue to digital processing techniques has allowed many new types of dynamics processing to be introduced. Also, in the example of a gain variation, a complex analogue multiplier is replaced by a much more straightforward and accurate digital multiplier. However, digital processing has also brought some new problems which were not present with previous analogue systems.

Consider, for example, a dynamics processor which relies on the detection of the peak level of an audio signal. Figure 1 of the accompanying drawings schematically illustrates such a processor in which an input digital audio signal (for example, a digital audio signal sampled at a 48kHz sampling rate to 16 bit resolution) is supplied in parallel to a multiplier 10 and to a processing chain 20.

The processing chain 20 comprises a peak (or envelope) detector 30, a logarithmic (linear to decibel) converter 40, a dynamics processing element 50, a control circuit 60, and a logarithmic (decibel to linear) converter 70.

The output of the logarithmic converter 70 is supplied as a second multiplicand to the multiplier 10, to be multiplied by sample values of the input digital audio signal. In this way, the output of the processing chain 20 provides a gain control for the input digital audio signal.

N 703 5 is 2 The dynamics processing element 50 operates in the logarithmic domain, i. e.

it receives envelope values and generates gain control values in decibels rather than as linear measures. This is so that the time constants, control values and other constants used by the dynamics processing element relate to a decibel law directly, which in turn makes the implementation of the dynamics processing element more simple and intuitive.

The dynamics processor of Figure 1 may be arranged to provide various different dynamics processing functions, depending on the way in which the dynamics processing element 50 generates an output gain control value in response to the detected envelope of the input digital audio signal. For the present explanation, consider the simple example whereby the dynamics processing element is arranged as a threshold compressor to provide an output of "0 decibels" (no compression) when the detected input envelope is under, say, -20dB and to provide an output indicative of, say, a 4:1 compression for higher detected input envelopes. (In general, the threshold level and the amount of applied compression will be selectable by the user under the control of the control circuit 60). This example compression response is illustrated schematically in Figure 2 of the accompanying drawings.

The gain control value output by the dynamics processing element 50 is then converted to a linear control value by the logarithmic converter 70 and supplied as a multiplicand to the multiplier 10.

In a discrete sampled system the peak value of the signal cannot be detected accurately simply by the envelope detector 30 evaluating the peak magnitude of successive sample. Sampling theory dictates that the actual representation of any sampled waveform is available only after suitable bandlimiting filtering to remove components above the Nyquist frequency.

This problem is illustrated schematically in Figures 3a and 3b of the accompanying drawings. These Figures show the magnitude of a 12 kHz audio signal, synchronised with sampling points (vertical lines) of a 48 kHz sampling rate. Time is represented along a horizontal axis.

In Figure 3a, the phase of the 12 kHz signal is such that peaks of that signal coincide with sampling points, so that a peak detector would detect the amplitude of the 12 kHz signal.

N 703 5 3 However, in Figure 3b, an otherwise similar 12 kHz signal is out of phase with the sampling frequency by 45'. In this case the maximum level of each sample (as detected by the simple peak detector mentioned above) will not be greater than cos(45') or 0.707 of the amplitude of the 12 kHz signal. Tlerefore, the simple peak detector can potentially introduce an error of up to MB depending on the relative phase of the sampled signal with respect to the sampling frequency.

In practice, significant errors of this type would occur where the signal is at or near integer divisions of the sample rate. In a real signal there is no synchronisation between the signal and the sampling rate, so this effect will manifest itself in aliasing errors, where the envelope (peak) detection is modulated by the difference between the sampled signal and the closest integer division of the sampling rate.

For example, if the signal frequency is 12.1 kHz with a sampling frequency of 48 kHz, there will be a 30% modulation of the detected envelope value by a 100 Hz signal - a modulation which is very hard to filter out because of its low frequency.

This kind of modulation, being permanently present, can cause very significant (and unwanted) distortion in dynamics processing. The situation may be improved by the use of large time constants, but this would make it impossible to achieve a fast response from the dynamics processor. Alternatively a large hysteresis in the dynamics processing controlled by the peak detection could be used, but this would lead to potentially large gain errors in the resulting processing.

This invention provides digital audio processing apparatus comprising:

an oversampler for oversampling an input digital audio signal to derive intermediate sample values at temporal positions between successive samples of the input digital audio signal; an envelope detector for detecting an envelope signal in response to instantaneous peak sample values of at least the intermediate samples; and a signal processor for applying a processing operation to the samples of the input digital audio signal in response to the detected envelope signal.

The invention recognises that the problems described above could be alleviated by simply increasing the sampling rate of the whole digital audio system in which the dynamics processor is used, but that this would dramatically increase the overall costs N 703 5 is 4 of generating and using digital audio signals. Therefore, the invention addresses the aliasing problems described above by the envelope detector generating intermediate sample values for use in detecting the envelope of the input digital audio signal. The detected envelope signal is used to control a dynamics processing function which is then applied to the input (i.e. non-oversampled) digital audio signal.

Thus, the envelope detection can be performed with increased accuracy by reference to the oversampled sample values, but these oversampled values do not directly form part of the processed digital audio signal and so do not have to be generated with the accuracy normally associated with digital processing directly applied to the signal chain.

The envelope signal may be detected by comparison of the oversampled values only, or by comparison of the oversampled values and the "real" samples of the input digital audio signal. If only the oversampled values are compared, then the oversampler's output level does not have to be accurately matched to the level of the real samples; it is, however, preferable that the oversampler's response is well matched so that the output intermediate sample values provide an accurate comparison of signal level.

Preferably, the oversampler is operable to generate an even number of intermediate sample values between successive samples of the input digital audio signal, which are preferably generated at sample positions substantially uniformly temporally spaced between successive samples of the input digital audio signal. In this way, filters used to generate the intermediate sample values may be well matched to one another, since the special case provided by an odd order finite impulse response filter centred on a mid coefficient of 'value 1 is avoided.

This invention also provides a method of digital audio processing, the method comprising the steps of.

oversampling an input digital audio signal to derive intermediate sample values at temporal positions between successive samples of the input digital audio signal; detecting an envelope signal in response to instantaneous peak sample values of at least the intermediate samples; and applying a processing operation to the samples of the input digital audio signal in response to the detected envelope signal.

N 703 5 Embodiments of the invention will now be described by way of example only with reference to the accompanying drawings in which:

Figure 1 is a schematic diagram of a dynamics processor; Figure 2 is a schematic graph of a simple compression response of the processor of Figure 1; Figures 3a and 3b are schematic diagrams illustrating potential aliasing problems; Figure 4 is a schematic diagram of an envelope detector according to an embodiment of the present invention; and Figure 5 is a schematic diagram illustrating the temporal positions of intermediate sample values generated by the detector of Figure 4, with respect to foreal" sampling positions of the input digital audio signal.

Figure 4 is a schematic diagram of an envelope detector according to an embodiment of the present invention.

The detector comprises a bank of four finite impulse response (FIR) filters 100, 110, 120 and 130, supplied with successively delayed samples from a series of delay elements 140.

Each of the four FIR filters comprises six multipliers 150 which multiply the respectively delayed sample value by a filter coefficient (shown schematically in Figure 4 as a shaded box) 160. The delayed sample values multiplied by their respective filter coefficients are added together by a series of adders 170 to generate a filtered output value. Thus, each of the FIR filters 100... 130 operates according to well-established filtering techniques.

The output of each of the four filters is converted to an absolute value by a respective absolute value converter 180. Then, pairs of the absolute values are compared by a first stage of maximum detectors 190, each of which provides an output equal to the greater of its respective two inputs. Finally, the output of the first stage maximum detectors 190 are supplied to a further maximum detector 200 to generate an output 210 representing the greatest of the four absolute values generated by the absolute value converters 180.

Thus, the peak value of the input digital signal is assessed by (a) deriving four intermediate sample values between successive "real" samples of the input digital N 703 5 6 signal; and (b) detecting the maximum of those intermediate sample values.

Figure 5 is a schematic diagram illustrating the temporal positions of the intermediate sample values generated by the dctector of Figure 4, with respect to IrreaV sampling positions of the digital audio signal.

Referring to Figures 4 and 5, each of the FIR filters 100... 130 has a set of filter coefficients selected (by conventional coefficient selection techniques) to give an output sample value at a particular temporal offset with respect to surrounding qfreal" samples. These temporal offsets are listed (in terms of the temporal spacing of two successive real samples) in the following table:

FIR filter temporal reference on Figure offset 5 filter 100 0.125 300 fitter 110 0.375 310 filter 120 0.625 320 filter 130 0.875 330 is N 703 5 1. Digital audio processing apparatus comprising: an oversampler for oversampling an input digital audio signal to derive intermediate sample values at temporal positions between successive samples of the input digital audio signal; an envelope detector for detecting an envelope signal in response to instantaneous peak sample values of at least the intermediate samples; and a signal processor for applying a processing operation to the samples of the input digital audio signal in response to the detected envelope signal.

7

Claims

2. Apparatus according to claim 1, in which the oversampler comprises one or more digital filters for deriving the intermediate sample values from respective groups of samples of the input digital audio signal.
3. Apparatus according to claim 1 or claim 2, in which the oversampler is operable to generate an even number of intermediate sample values between successive samples of the input digital audio signal.
4. Apparatus according to claim 3, in which the intermediate sample values are generated at sample positions substantially uniformly temporally spaced between successive samples of the input digital audio signal.
5.

A method of digital audio processing, the method comprising the steps of:

oversampling an input digital audio signal to derive intermediate sample values at temporal positions between successive samples of the input digital audio signal; detecting an envelope signal in response to instantaneous peak sample values of at least the intermediate samples; and applying a processing operation to the samples of the input digital audio signal in response to the detected envelope signal.
6. Digital audio processing apparatus substantially as hereinbefore described with N 703 5 8 reference to the accompanying drawings.
7. A method of digital audio processing, the method being substantially as hereinbefore described with reference to the accompanying drawings.