Author: superdump
Date: Sat Jul 12 22:30:18 2008
New Revision: 2771
Log:
Multiply samples by mul_bias before converting to 16-bit int to bring them to
the correct range
Modified:
amr/amrnbfloatdec.c
Modified: amr/amrnbfloatdec.c
==============================================================================
--- amr/amrnbfloatdec.c (original)
+++ amr/amrnbfloatdec.c Sat Jul 12 22:30:18 2008
@@ -1297,7 +1297,7 @@ static int amrnb_decode_frame(AVCodecCon
// convert float samples to 16-bit integer
for(i=0; i<AMR_SUBFRAME_SIZE; i++) {
- p->samples_in[i] += p->add_bias;
+ p->samples_in[i] = p->samples_in[i] * p->mul_bias + p->add_bias;
}
p->dsp.float_to_int16(p->samples_out, p->samples_in,
AMR_SUBFRAME_SIZE);
memcpy(&buf_out[subframe*AMR_SUBFRAME_SIZE], p->samples_out,
AMR_SUBFRAME_SIZE*sizeof(int16_t));
_______________________________________________
FFmpeg-soc mailing list
[email protected]
https://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-soc
