generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

aacenc: Fix a rounding bug in aacpsy channel bitrate computation

Browse Source 
Signed-off-by: Martin Storsjö <martin@martin.st>
This commit is contained in:
Claudio Freire 2013-05-04 18:35:49 -03:00 committed by Martin Storsjö
parent d10499a32a
commit 7c71ada4ca
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
libavcodec/aacpsy.c
View File

@ -312,7 +312,7 @@ static av_cold int psy_3gpp_init(FFPsyContext *ctx) {
AacPsyCoeffs *coeffs = pctx->psy_coef[j];
const uint8_t *band_sizes = ctx->bands[j];
float line_to_frequency = ctx->avctx->sample_rate / (j ? 256.f : 2048.0f);
float avg_chan_bits = chan_bitrate / ctx->avctx->sample_rate * (j ? 128.0f : 1024.0f);
float avg_chan_bits = chan_bitrate * (j ? 128.0f : 1024.0f) / ctx->avctx->sample_rate;
/* reference encoder uses 2.4% here instead of 60% like the spec says */
float bark_pe = 0.024f * PSY_3GPP_BITS_TO_PE(avg_chan_bits) / num_bark;
float en_spread_low = j ? PSY_3GPP_EN_SPREAD_LOW_S : PSY_3GPP_EN_SPREAD_LOW_L;