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
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
|
/*
* This file is part of FFmpeg.
*
* FFmpeg 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.
*
* FFmpeg 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 FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#undef ONE
#undef ftype
#undef SAMPLE_FORMAT
#if DEPTH == 32
#define SAMPLE_FORMAT float
#define ftype float
#define ONE 1.f
#else
#define SAMPLE_FORMAT double
#define ftype double
#define ONE 1.0
#endif
#define fn3(a,b) a##_##b
#define fn2(a,b) fn3(a,b)
#define fn(a) fn2(a, SAMPLE_FORMAT)
#if DEPTH == 64
static double scalarproduct_double(const double *v1, const double *v2, int len)
{
double p = 0.0;
for (int i = 0; i < len; i++)
p += v1[i] * v2[i];
return p;
}
#endif
static ftype fn(fir_sample)(AudioNLMSContext *s, ftype sample, ftype *delay,
ftype *coeffs, ftype *tmp, int *offset)
{
const int order = s->order;
ftype output;
delay[*offset] = sample;
memcpy(tmp, coeffs + order - *offset, order * sizeof(ftype));
#if DEPTH == 32
output = s->fdsp->scalarproduct_float(delay, tmp, s->kernel_size);
#else
output = scalarproduct_double(delay, tmp, s->kernel_size);
#endif
if (--(*offset) < 0)
*offset = order - 1;
return output;
}
static ftype fn(process_sample)(AudioNLMSContext *s, ftype input, ftype desired,
ftype *delay, ftype *coeffs, ftype *tmp, int *offsetp)
{
const int order = s->order;
const ftype leakage = s->leakage;
const ftype mu = s->mu;
const ftype a = ONE - leakage;
ftype sum, output, e, norm, b;
int offset = *offsetp;
delay[offset + order] = input;
output = fn(fir_sample)(s, input, delay, coeffs, tmp, offsetp);
e = desired - output;
#if DEPTH == 32
sum = s->fdsp->scalarproduct_float(delay, delay, s->kernel_size);
#else
sum = scalarproduct_double(delay, delay, s->kernel_size);
#endif
norm = s->eps + sum;
b = mu * e / norm;
if (s->anlmf)
b *= e * e;
memcpy(tmp, delay + offset, order * sizeof(ftype));
#if DEPTH == 32
s->fdsp->vector_fmul_scalar(coeffs, coeffs, a, s->kernel_size);
s->fdsp->vector_fmac_scalar(coeffs, tmp, b, s->kernel_size);
#else
s->fdsp->vector_dmul_scalar(coeffs, coeffs, a, s->kernel_size);
s->fdsp->vector_dmac_scalar(coeffs, tmp, b, s->kernel_size);
#endif
memcpy(coeffs + order, coeffs, order * sizeof(ftype));
switch (s->output_mode) {
case IN_MODE: output = input; break;
case DESIRED_MODE: output = desired; break;
case OUT_MODE: output = desired - output; break;
case NOISE_MODE: output = input - output; break;
case ERROR_MODE: break;
}
return output;
}
static int fn(filter_channels)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
{
AudioNLMSContext *s = ctx->priv;
AVFrame *out = arg;
const int start = (out->ch_layout.nb_channels * jobnr) / nb_jobs;
const int end = (out->ch_layout.nb_channels * (jobnr+1)) / nb_jobs;
for (int c = start; c < end; c++) {
const ftype *input = (const ftype *)s->frame[0]->extended_data[c];
const ftype *desired = (const ftype *)s->frame[1]->extended_data[c];
ftype *delay = (ftype *)s->delay->extended_data[c];
ftype *coeffs = (ftype *)s->coeffs->extended_data[c];
ftype *tmp = (ftype *)s->tmp->extended_data[c];
int *offset = (int *)s->offset->extended_data[c];
ftype *output = (ftype *)out->extended_data[c];
for (int n = 0; n < out->nb_samples; n++) {
output[n] = fn(process_sample)(s, input[n], desired[n], delay, coeffs, tmp, offset);
if (ctx->is_disabled)
output[n] = input[n];
}
}
return 0;
}
|
