From f7700bb398469858fdae3c6d0be22ff84d157ab3 Mon Sep 17 00:00:00 2001 From: JIe Jie Date: Mon, 26 Feb 2024 09:10:10 +0800 Subject: [PATCH] =?UTF-8?q?fftw3=20=E5=82=85=E9=87=8C=E5=8F=B6=E5=8F=98?= =?UTF-8?q?=E6=8D=A2=E6=B1=82=E9=A2=91=E7=8E=87?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- src/audioDecoder.cc | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/src/audioDecoder.cc b/src/audioDecoder.cc index ce547c7..2b84f59 100644 --- a/src/audioDecoder.cc +++ b/src/audioDecoder.cc @@ -1,10 +1,12 @@ #include "audioDecoder.h" +#include #include #include extern "C" { #include "libswresample/swresample.h" +#include "fftw3.h" } int RequestAudioFrame(AudioParam& param, uint8_t* audioBuffer, int bufSize) @@ -37,6 +39,25 @@ int RequestAudioFrame(AudioParam& param, uint8_t* audioBuffer, int bufSize) if (frame->ch_layout.nb_channels > 0) { av_channel_layout_default(&frame->ch_layout, frame->ch_layout.nb_channels); } + int nbSamples = frame->nb_samples; + uint8_t* audioData = frame->data[0]; + fftw_complex* fftwInput = static_cast(fftw_malloc(sizeof(fftw_complex) * frame->nb_samples)); + fftw_complex* fftwOutput = static_cast(fftw_malloc(sizeof(fftw_complex) * frame->nb_samples)); + + for (int i = 0; i < nbSamples; i++) + { + fftwInput[i][0] = audioData[i]; + fftwInput[i][1] = .0f; + } + fftw_plan fftwPlan = fftw_plan_dft_1d(nbSamples, fftwInput, fftwOutput, FFTW_FORWARD, FFTW_ESTIMATE); + fftw_execute(fftwPlan); + + for (int i = 0; i < nbSamples; i++) + { + double magnitude = sqrt(fftwOutput[i][0] * fftwOutput[i][0] + fftwOutput[i][1] * fftwOutput[i][1]); + double phase = atan2(fftwOutput[i][1], fftwOutput[i][0]); + std::cout << "mag: " << magnitude << "\n phase: " << phase << "\n"; + } constexpr AVSampleFormat dstFormat = AV_SAMPLE_FMT_S16; swr_alloc_set_opts2(&swrCtx, &frame->ch_layout, dstFormat, frame->sample_rate, &frame->ch_layout, static_cast(frame->format), frame->sample_rate, 0, nullptr);