【鼎革‧革鼎】︰ Raspbian Stretch 《六之 K.3-言語界面-7.2B 》

就算已經打通了『檔案介面』︰

 

通常最好還是先細讀文件︰

recognizer_instance.dynamic_energy_threshold = True # type: bool

Represents whether the energy level threshold (see recognizer_instance.energy_threshold) for sounds should be automatically adjusted based on the currently ambient noise level while listening. Can be changed.

Recommended for situations where the ambient noise level is unpredictable, which seems to be the majority of use cases. If the ambient noise level is strictly controlled, better results might be achieved by setting this to False to turn it off.

recognizer_instance.dynamic_energy_adjustment_damping = 0.15 # type: float

If the dynamic energy threshold setting is enabled (see recognizer_instance.dynamic_energy_threshold), represents approximately the fraction of the current energy threshold that is retained after one second of dynamic threshold adjustment. Can be changed (not recommended).

Lower values allow for faster adjustment, but also make it more likely to miss certain phrases (especially those with slowly changing volume). This value should be between 0 and 1. As this value approaches 1, dynamic adjustment has less of an effect over time. When this value is 1, dynamic adjustment has no effect.

recognizer_instance.dynamic_energy_adjustment_ratio = 1.5 # type: float

If the dynamic energy threshold setting is enabled (see recognizer_instance.dynamic_energy_threshold), represents the minimum factor by which speech is louder than ambient noise. Can be changed (not recommended).

For example, the default value of 1.5 means that speech is at least 1.5 times louder than ambient noise. Smaller values result in more false positives (but fewer false negatives) when ambient noise is loud compared to speech.

recognizer_instance.pause_threshold = 0.8 # type: float

Represents the minimum length of silence (in seconds) that will register as the end of a phrase. Can be changed.

Smaller values result in the recognition completing more quickly, but might result in slower speakers being cut off.

 

不必急於七測八試!避免冒闖『概念』叢林?

Ambient noise level

In atmospheric sounding and noise pollution, ambient noise level (sometimes called background noise level, reference sound level, or room noise level) is the background sound pressure level at a given location, normally specified as a reference level to study a new intrusive sound source.

Ambient sound levels are often measured in order to map sound conditions over a spatial regime to understand their variation with locale. In this case the product of the investigation is a sound level contour map. Alternatively ambient noise levels may be measured to provide a reference point for analyzing an intrusive sound to a given environment. For example, sometimes aircraft noise is studied by measuring ambient sound without presence of any overflights, and then studying the noise addition by measurement or computer simulation of overflight events. Or roadway noise is measured as ambient sound, prior to introducing a hypothetical noise barrier intended to reduce that ambient noise level.

Ambient noise level is measured with a sound level meter. It is usually measured in dB relative to a reference pressure of 0.00002 Pa, i.e., 20 μPa (micropascals) in SI units. A pascal is a newton per square meter. The centimeter-gram-second system of units, the reference sound pressure for measuring ambient noise level is 0.0002 dyn/cm2. Most frequently ambient noise levels are measured using a frequency weighting filter, the most common being the A-weighting scale, such that resulting measurements are denoted dB(A), or decibels on the A-weighting scale.

Integrating Sound Level Meter

……

Sound level meter

A sound level meter is used for acoustic (sound that travels through air) measurements. It is commonly a hand-held instrument with a microphone. The diaphragm of the microphone responds to changes in air pressure caused by sound waves. That is why the instrument is sometimes referred to as a Sound Pressure Level (SPL) Meter. This movement of the diaphragm, i.e. the sound pressure deviation (pascal Pa), is converted into an electrical signal (volts V).

A microphone is distinguishable by the voltage value produced when a known, constant sound pressure is applied. This is known as the microphone sensitivity. The instrument needs to know the sensitivity of the particular microphone being used. Using this information, the instrument is able to accurately convert the electrical signal back to a sound pressure, and display the resulting sound pressure level (decibels dB SPL).

Sound level meters are commonly used in noise pollution studies for the quantification of different kinds of noise, especially for industrial, environmental and aircraft noise. The current international standard that specifies sound level meter functionality and performances is the IEC 61672-1:2013. However, the reading from a sound level meter does not correlate well to human-perceived loudness, which is better measured by a loudness meter. Specific loudness is a compressive nonlinearity that depends on level and also frequency, which can be calculated in a number of different ways.[1]

An integrating-averaging Cirrus Optimus which complies with IEC 61672-1:2002

………

Measurements

Frequency weighting

‘Pattern Approved’ sound level meters offer noise measurements with A, C and Z frequency weighting.[5]

Z-Weighting represents the actual sound produced. A-Weighting, with less lower and higher frequencies, represents what humans are capable of hearing. C-Weighting, more sensitive to the lower frequencies, represents what humans hear when the sound is loud.

The IEC 61672-1:2013 mandates the inclusion of an A-frequency-weighting filter in all sound level meters, and also describes C and Z (zero) frequency weightings. The older B and D frequency weightings are now obsolete and are no longer described in the standard.

In almost all countries, the use of A-frequency-weighting is mandated to be used for the protection of workers against noise-induced hearing loss. The A-frequency curve was based on the historical equal-loudness contours and while arguably A-frequency-weighting is no longer the ideal frequency weighting on purely scientific grounds, it is nonetheless the legally required standard for almost all such measurements and has the huge practical advantage that old data can be compared with new measurements. It is for these reasons that A-frequency-weighting is the only weighting mandated by the international standard, the frequency weightings ‘C’ and ‘Z’ being optional fitments.

Originally, the A-frequency-weighting was only meant for quiet sounds in the region of 40 dB sound pressure level (SPL), but is now mandated for all levels. C-frequency weighting is however still used in the measurement of the peak value of a noise in some legislation, but B-frequency weighting – a halfway house between ‘A’ and ‘C’ has almost no practical use. D-frequency-weighting was designed for use in measuring aircraft noise when non-bypass jets were being measured and after the demise of Concord, these are all military types. For all civil aircraft noise measurements, A-frequency-weighting is used as is mandated by the ISO and ICAO standards.

A, C and Z frequency weightings for sound

 

回顧匯聚已知之事??

久居都市早已習慣眼不見星空,耳難得清靜,故爾雖非不知

DoA on ReSpeaker 4-Mic Array for Raspberry Pi

With DoA(Direction of Arrial), ReSpeaker 4-Mic Array is able to find the direction where the sound source is located.

………

 

之設計目的,惟願先講『響度計』呦◎

Ebumeter – Quick guide

Ebumeter provides level metering according to the EBU R-128 recommendation. The current release implements all features required by the EBU document except the oversampled peak level monitoring. This will be added in a future release.

For more detailed info on R-128 see the EBU site or the papers section of my website.

The upper bargraph shows either the Momentary or the Short term loudness as selected by the M and S buttons to the right. The two thinner ones below display the Loudness range (LRA) and the Integrated loudness (I) which are also shown in numerical form.

The +9 and +18 buttons switch between the two ranges required for an EBU-mode meter. The LU and FS buttons select either the relative scale in LU or the absolute one in LUFS.

The Stop, Start and Reset buttons control the Integrated loudness measurement.

─── 摘自《【鼎革‧革鼎】︰Raspbian Stretch 《六之 I‧導言 》

 

重新梳理尚未讀或不解的術語!!

 

Papers and technotes

Audio measurements workshop
Fons Adriaensen
LAC2014 Slides (pdf, 177k)

Some things you need to know when doing audio measurements.

………

 

 

 

 

 

 

 

發表迴響

你的電子郵件位址並不會被公開。 必要欄位標記為 *