various functions useful in audio signal processing More...

Functions
fvec_t *	new_aubio_window (char_t *window_type, uint_t size)
	create window
uint_t	fvec_set_window (fvec_t window, char_t window_type)
	set elements of a vector to window coefficients
smpl_t	aubio_unwrap2pi (smpl_t phase)
	compute the principal argument
smpl_t	aubio_bintomidi (smpl_t bin, smpl_t samplerate, smpl_t fftsize)
	convert frequency bin to midi value
smpl_t	aubio_miditobin (smpl_t midi, smpl_t samplerate, smpl_t fftsize)
	convert midi value to frequency bin
smpl_t	aubio_bintofreq (smpl_t bin, smpl_t samplerate, smpl_t fftsize)
	convert frequency bin to frequency (Hz)
smpl_t	aubio_freqtobin (smpl_t freq, smpl_t samplerate, smpl_t fftsize)
	convert frequency (Hz) to frequency bin
smpl_t	aubio_hztomel (smpl_t freq)
	convert frequency (Hz) to mel
smpl_t	aubio_meltohz (smpl_t mel)
	convert mel to frequency (Hz)
smpl_t	aubio_hztomel_htk (smpl_t freq)
	convert frequency (Hz) to mel
smpl_t	aubio_meltohz_htk (smpl_t mel)
	convert mel to frequency (Hz)
smpl_t	aubio_freqtomidi (smpl_t freq)
	convert frequency (Hz) to midi value (0-128)
smpl_t	aubio_miditofreq (smpl_t midi)
	convert midi value (0-128) to frequency (Hz)
void	aubio_cleanup (void)
	clean up cached memory at the end of program
smpl_t	aubio_zero_crossing_rate (fvec_t *v)
	zero-crossing rate (ZCR)
smpl_t	aubio_level_lin (const fvec_t *v)
	compute sound level on a linear scale
smpl_t	aubio_db_spl (const fvec_t *v)
	compute sound pressure level (SPL) in dB
uint_t	aubio_silence_detection (const fvec_t *v, smpl_t threshold)
	check if buffer level in dB SPL is under a given threshold
smpl_t	aubio_level_detection (const fvec_t *v, smpl_t threshold)
	get buffer level if level >= threshold, 1.
void	fvec_clamp (fvec_t *in, smpl_t absmax)
	clamp the values of a vector within the range [-abs(max), abs(max)]

Detailed Description

various functions useful in audio signal processing

Definition in file musicutils.h.

Function Documentation

◆ aubio_bintofreq()

smpl_t aubio_bintofreq	(	smpl_t	bin,
		smpl_t	samplerate,
		smpl_t	fftsize )

convert frequency bin to frequency (Hz)

References aubio_bintofreq().

Referenced by aubio_bintofreq().

◆ aubio_bintomidi()

smpl_t aubio_bintomidi	(	smpl_t	bin,
		smpl_t	samplerate,
		smpl_t	fftsize )

convert frequency bin to midi value

References aubio_bintomidi().

Referenced by aubio_bintomidi().

◆ aubio_cleanup()

void aubio_cleanup ( void )

clean up cached memory at the end of program

This function should be used at the end of programs to purge all cached memory. So far it is only useful to clean FFTW's cache.

Examples: onset/test-onset.c, pitch/test-pitch.c, spectral/test-fft.c, spectral/test-filterbank.c, spectral/test-filterbank_mel.c, spectral/test-mfcc.c, spectral/test-phasevoc.c, spectral/test-specdesc.c, spectral/test-tss.c, synth/test-sampler.c, synth/test-wavetable.c, tempo/test-tempo.c, and temporal/test-filter.c.

References aubio_cleanup().

Referenced by aubio_cleanup().

◆ aubio_db_spl()

smpl_t aubio_db_spl ( const fvec_t * v )

compute sound pressure level (SPL) in dB

This quantity is often wrongly called 'loudness'.

This gives ten times the log10 of the average of the square amplitudes.

Parameters

v	vector to compute dB SPL from

Returns: level of v in dB SPL

References aubio_db_spl().

Referenced by aubio_db_spl().

◆ aubio_freqtobin()

smpl_t aubio_freqtobin	(	smpl_t	freq,
		smpl_t	samplerate,
		smpl_t	fftsize )

convert frequency (Hz) to frequency bin

References aubio_freqtobin().

Referenced by aubio_freqtobin().

◆ aubio_freqtomidi()

smpl_t aubio_freqtomidi ( smpl_t freq )

convert frequency (Hz) to midi value (0-128)

References aubio_freqtomidi().

Referenced by aubio_freqtomidi().

◆ aubio_hztomel()

smpl_t aubio_hztomel ( smpl_t freq )

convert frequency (Hz) to mel

Parameters

freq	input frequency, in Hz

Returns: output mel

Converts a scalar from the frequency domain to the mel scale using Slaney Auditory Toolbox's implementation:

If \( f < 1000 \), \( m = 3 f / 200 \).

If \( f >= 1000 \), \( m = 1000 + 27 \frac{{ln}(f) - ln(1000))} {{ln}(6400) - ln(1000)} \)

◆ aubio_hztomel_htk()

smpl_t aubio_hztomel_htk ( smpl_t freq )

convert frequency (Hz) to mel

Parameters

freq	input frequency, in Hz

Returns: output mel

Converts a scalar from the frequency domain to the mel scale, using the equation defined by O'Shaughnessy, as implemented in the HTK speech recognition toolkit:

\( m = 1127 + ln(1 + \frac{f}{700}) \)

References

Douglas O'Shaughnessy (1987). Speech communication: human and machine. Addison-Wesley. p. 150. ISBN 978-0-201-16520-3.

HTK Speech Recognition Toolkit: http://htk.eng.cam.ac.uk/

References aubio_hztomel_htk().

Referenced by aubio_hztomel_htk().

◆ aubio_level_detection()

smpl_t aubio_level_detection	(	const fvec_t *	v,
		smpl_t	threshold )

get buffer level if level >= threshold, 1.

otherwise

Parameters

v	vector to get level from
threshold	threshold in dB SPL

Returns: level in dB SPL if level >= threshold, 1. otherwise

References aubio_level_detection().

Referenced by aubio_level_detection().

◆ aubio_level_lin()

smpl_t aubio_level_lin ( const fvec_t * v )

compute sound level on a linear scale

This gives the average of the square amplitudes.

Parameters

v	vector to compute level from

Returns: level of v

Examples: examples/aubiopitch.c.

References aubio_level_lin().

Referenced by aubio_level_lin().

◆ aubio_meltohz()

smpl_t aubio_meltohz ( smpl_t mel )

convert mel to frequency (Hz)

Parameters

mel input mel

Returns: output frequency, in Hz

Converts a scalar from the mel scale to the frequency domain using Slaney Auditory Toolbox's implementation:

If \( f < 1000 \), \( f = 200 m/3 \).

If \( f \geq 1000 \), \( f = 1000 + \left(\frac{6400}{1000}\right) ^{\frac{m - 1000}{27}} \)

References

Malcolm Slaney, Auditory Toolbox Version 2, Technical Report #1998-010 https://engineering.purdue.edu/~malcolm/interval/1998-010/

References aubio_meltohz().

Referenced by aubio_meltohz().

◆ aubio_meltohz_htk()

smpl_t aubio_meltohz_htk ( smpl_t mel )

convert mel to frequency (Hz)

Parameters

mel input mel

Returns: output frequency, in Hz

Converts a scalar from the mel scale to the frequency domain, using the equation defined by O'Shaughnessy, as implemented in the HTK speech recognition toolkit:

\( f = 700 * {e}^\left(\frac{f}{1127} - 1\right) \)

◆ aubio_miditobin()

smpl_t aubio_miditobin	(	smpl_t	midi,
		smpl_t	samplerate,
		smpl_t	fftsize )

convert midi value to frequency bin

References aubio_miditobin().

Referenced by aubio_miditobin().

◆ aubio_miditofreq()

smpl_t aubio_miditofreq ( smpl_t midi )

convert midi value (0-128) to frequency (Hz)

References aubio_miditofreq().

Referenced by aubio_miditofreq().

◆ aubio_silence_detection()

uint_t aubio_silence_detection	(	const fvec_t *	v,
		smpl_t	threshold )

check if buffer level in dB SPL is under a given threshold

Parameters

v	vector to get level from
threshold	threshold in dB SPL

Returns: 0 if level is under the given threshold, 1 otherwise

Examples: examples/aubiotrack.c.

References aubio_silence_detection().

Referenced by aubio_silence_detection().

◆ aubio_unwrap2pi()

smpl_t aubio_unwrap2pi ( smpl_t phase )

compute the principal argument

This function maps the input phase to its corresponding value wrapped in the range \( [-\pi, \pi] \).

Parameters

phase unwrapped phase to map to the unit circle

Returns: equivalent phase wrapped to the unit circle

References aubio_unwrap2pi().

Referenced by aubio_unwrap2pi().

◆ aubio_zero_crossing_rate()

smpl_t aubio_zero_crossing_rate ( fvec_t * v )

zero-crossing rate (ZCR)

The zero-crossing rate is the number of times a signal changes sign, divided by the length of this signal.

Parameters

v	vector to compute ZCR from

Returns: zero-crossing rate of v

References aubio_zero_crossing_rate().

Referenced by aubio_zero_crossing_rate().

◆ fvec_clamp()

void fvec_clamp	(	fvec_t *	in,
		smpl_t	absmax )

clamp the values of a vector within the range [-abs(max), abs(max)]

Parameters

in	vector to clamp
absmax	maximum value over which input vector elements should be clamped

References fvec_clamp().

Referenced by fvec_clamp().

◆ fvec_set_window()

uint_t fvec_set_window	(	fvec_t *	window,
		char_t *	window_type )

set elements of a vector to window coefficients

Parameters

window	exsting fvec_t to use
window_type	type of the window to create

List of available window types: "rectangle", "hamming", "hanning", "hanningz", "blackman", "blackman_harris", "gaussian", "welch", "parzen", "default".

"default" is equivalent to "hanningz".

References:

Window function on Wikipedia
Amalia de Götzen, Nicolas Bernardini, and Daniel Arfib. Traditional (?) implementations of a phase vocoder: the tricks of the trade. In Proceedings of the International Conference on Digital Audio Effects (DAFx-00), pages 37–44, Uni- versity of Verona, Italy, 2000. (pdf)

References fvec_set_window().

Referenced by fvec_set_window().

◆ new_aubio_window()

fvec_t * new_aubio_window	(	char_t *	window_type,
		uint_t	size )

create window

Parameters

window_type	type of the window to create
size	length of the window to create (see fvec_set_window())

References new_aubio_window().

Referenced by new_aubio_window().

Functions

Detailed Description

Function Documentation

◆ aubio_bintofreq()

◆ aubio_bintomidi()

◆ aubio_cleanup()

◆ aubio_db_spl()

◆ aubio_freqtobin()

◆ aubio_freqtomidi()

◆ aubio_hztomel()

See also

◆ aubio_hztomel_htk()

See also

References

◆ aubio_level_detection()

◆ aubio_level_lin()

◆ aubio_meltohz()

See also

References

◆ aubio_meltohz_htk()

See also

◆ aubio_miditobin()

◆ aubio_miditofreq()

◆ aubio_silence_detection()

◆ aubio_unwrap2pi()

◆ aubio_zero_crossing_rate()

◆ fvec_clamp()

◆ fvec_set_window()

◆ new_aubio_window()