1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112

/*
* TwoLAME: an optimized MPEG Audio Layer Two encoder
*
* Copyright (C) 20012004 Michael Cheng
* Copyright (C) 20042006 The TwoLAME Project
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 021111307 USA
*
* $Id: ath.c 156 20070320 23:57:35Z nhumfrey $
*
*/
#include <stdio.h>
#include <math.h>
#include "twolame.h"
#include "common.h"
#include "ath.h"
/* freq in hz */
FLOAT ath_db(FLOAT f, FLOAT value)
{
/* from Painter & Spanias
modified by Gabriel Bouvigne to better fit the reality
ath = 3.640 * pow(f,0.8)
 6.800 * exp(0.6*pow(f3.4,2.0))
+ 6.000 * exp(0.15*pow(f8.7,2.0))
+ 0.6* 0.001 * pow(f,4.0);
In the past LAME was using the Painter &Spanias formula.
But we had some recurrent problems with HF content.
We measured real ATH values, and found the older formula
to be inacurate in the higher part. So we made this new
formula and this solved most of HF problematic testcases.
The tradeoff is that in VBR mode it increases a lot the
bitrate.*/
/*this curve can be udjusted according to the VBR scale:
it adjusts from something close to Painter & Spanias
on V9 up to Bouvigne's formula for V0. This way the VBR
bitrate is more balanced according to the V value.*/
FLOAT ath;
FLOAT valueold = 0.0;
if (f < .3)
f=3410;
f /= 1000; // convert to khz
f = MAX(0.01, f);
f = MIN(18.0, f);
ath = 3.640 * pow(f,0.8)
 6.800 * exp(0.6*pow(f3.4,2.0))
+ 6.000 * exp(0.15*pow(f8.7,2.0))
+ (0.6+0.04*valueold)* 0.001 * pow(f,4.0);
/* MFC Feb 2003
I've changed the fudge technique on the code.
The "l [float]" value raises/lowers the ATH by this many dB */
return (ath + value);
}
/* Convert ATH values from dB into energy values as required by the psycho model */
FLOAT ath_energy(FLOAT freq, FLOAT value)
{
FLOAT db;
db = ath_db(freq, 0) + value; // Originally: ath_db(freq,value)
/* The values in the standard, and from the ATH formula are in dB.
In the psycho model we are working in the energy domain. Hence the values that
are in the absthr_X tables are not in dB. This function converts from dB into the energy domain.
As noted on the LAME mailing list from years ago (MFC FIX find the reference), the
absolute threhsold of hearing values in the tables in the standard are dodgy  the
ATH in the tables do not correspond to any previously known values of the ATH.
From ISO 11172 Tables D.4.x
"A value of 0dB represents a level in the absolute threshold calculation of
96dB below the energy of a sine wave of amplitude 32760."
But I still don't know why the factor of 41.837375 is the value that it is.
MFC Feb 2003
*/
return(pow(10.0, (db+41.837375)*0.1));
}
/* Convert a frequency (in Hz) to a bark value
Taken from LAME. MFC Feb 2003
see for example "Zwicker: Psychoakustik, 1982; ISBN 3540114017 */
FLOAT ath_freq2bark(FLOAT freq)
{
if (freq<0) freq=0;
freq = freq * 0.001;
return 13.0*atan(.76*freq) + 3.5*atan(freq*freq/(7.5*7.5));
}
// vim:ts=4:sw=4:nowrap:
