Recherche avancée
Autres articles (68)
-
MediaSPIP version 0.1 Beta
16 avril 2011, parMediaSPIP 0.1 beta est la première version de MediaSPIP décrétée comme "utilisable".
Le fichier zip ici présent contient uniquement les sources de MediaSPIP en version standalone.
Pour avoir une installation fonctionnelle, il est nécessaire d’installer manuellement l’ensemble des dépendances logicielles sur le serveur.
Si vous souhaitez utiliser cette archive pour une installation en mode ferme, il vous faudra également procéder à d’autres modifications (...) -
MediaSPIP 0.1 Beta version
25 avril 2011, parMediaSPIP 0.1 beta is the first version of MediaSPIP proclaimed as "usable".
The zip file provided here only contains the sources of MediaSPIP in its standalone version.
To get a working installation, you must manually install all-software dependencies on the server.
If you want to use this archive for an installation in "farm mode", you will also need to proceed to other manual (...) -
Amélioration de la version de base
13 septembre 2013Jolie sélection multiple
Le plugin Chosen permet d’améliorer l’ergonomie des champs de sélection multiple. Voir les deux images suivantes pour comparer.
Il suffit pour cela d’activer le plugin Chosen (Configuration générale du site > Gestion des plugins), puis de configurer le plugin (Les squelettes > Chosen) en activant l’utilisation de Chosen dans le site public et en spécifiant les éléments de formulaires à améliorer, par exemple select[multiple] pour les listes à sélection multiple (...)
Sur d’autres sites (8039)
-
C++ ffmpeg lib version 7.0 - export audio to different format
4 septembre 2024, par Chris PI want to make a C++ lib named cppdub which will mimic the python module pydub.


One main function is to export the AudioSegment to a file with a specific format (example : mp3).


The code is :



AudioSegment AudioSegment::from_file(const std::string& file_path, const std::string& format, const std::string& codec,
 const std::map& parameters, int start_second, int duration) {

 avformat_network_init();
 av_log_set_level(AV_LOG_ERROR); // Adjust logging level as needed

 AVFormatContext* format_ctx = nullptr;
 if (avformat_open_input(&format_ctx, file_path.c_str(), nullptr, nullptr) != 0) {
 std::cerr << "Error: Could not open audio file." << std::endl;
 return AudioSegment(); // Return an empty AudioSegment on failure
 }

 if (avformat_find_stream_info(format_ctx, nullptr) < 0) {
 std::cerr << "Error: Could not find stream information." << std::endl;
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 int audio_stream_index = -1;
 for (unsigned int i = 0; i < format_ctx->nb_streams; i++) {
 if (format_ctx->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
 audio_stream_index = i;
 break;
 }
 }

 if (audio_stream_index == -1) {
 std::cerr << "Error: Could not find audio stream." << std::endl;
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 AVCodecParameters* codec_par = format_ctx->streams[audio_stream_index]->codecpar;
 const AVCodec* my_codec = avcodec_find_decoder(codec_par->codec_id);
 AVCodecContext* codec_ctx = avcodec_alloc_context3(my_codec);

 if (!codec_ctx) {
 std::cerr << "Error: Could not allocate codec context." << std::endl;
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 if (avcodec_parameters_to_context(codec_ctx, codec_par) < 0) {
 std::cerr << "Error: Could not initialize codec context." << std::endl;
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 if (avcodec_open2(codec_ctx, my_codec, nullptr) < 0) {
 std::cerr << "Error: Could not open codec." << std::endl;
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 SwrContext* swr_ctx = swr_alloc();
 if (!swr_ctx) {
 std::cerr << "Error: Could not allocate SwrContext." << std::endl;
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }
 codec_ctx->sample_rate = 44100;
 // Set up resampling context to convert to S16 format with 2 bytes per sample
 av_opt_set_chlayout(swr_ctx, "in_chlayout", &codec_ctx->ch_layout, 0);
 av_opt_set_int(swr_ctx, "in_sample_rate", codec_ctx->sample_rate, 0);
 av_opt_set_sample_fmt(swr_ctx, "in_sample_fmt", codec_ctx->sample_fmt, 0);

 AVChannelLayout dst_ch_layout;
 av_channel_layout_copy(&dst_ch_layout, &codec_ctx->ch_layout);
 av_channel_layout_uninit(&dst_ch_layout);
 av_channel_layout_default(&dst_ch_layout, 2);

 av_opt_set_chlayout(swr_ctx, "out_chlayout", &dst_ch_layout, 0);
 av_opt_set_int(swr_ctx, "out_sample_rate", codec_ctx->sample_rate, 0); // Match input sample rate
 av_opt_set_sample_fmt(swr_ctx, "out_sample_fmt", AV_SAMPLE_FMT_S16, 0); // Force S16 format

 if (swr_init(swr_ctx) < 0) {
 std::cerr << "Error: Failed to initialize the resampling context" << std::endl;
 swr_free(&swr_ctx);
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 AVPacket packet;
 AVFrame* frame = av_frame_alloc();
 if (!frame) {
 std::cerr << "Error: Could not allocate frame." << std::endl;
 swr_free(&swr_ctx);
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 std::vector<char> output;
 while (av_read_frame(format_ctx, &packet) >= 0) {
 if (packet.stream_index == audio_stream_index) {
 if (avcodec_send_packet(codec_ctx, &packet) == 0) {
 while (avcodec_receive_frame(codec_ctx, frame) == 0) {
 if (frame->pts != AV_NOPTS_VALUE) {
 frame->pts = av_rescale_q(frame->pts, codec_ctx->time_base, format_ctx->streams[audio_stream_index]->time_base);
 }

 uint8_t* output_buffer;
 int output_samples = av_rescale_rnd(
 swr_get_delay(swr_ctx, codec_ctx->sample_rate) + frame->nb_samples,
 codec_ctx->sample_rate, codec_ctx->sample_rate, AV_ROUND_UP);

 int output_buffer_size = av_samples_get_buffer_size(
 nullptr, 2, output_samples, AV_SAMPLE_FMT_S16, 1);

 output_buffer = (uint8_t*)av_malloc(output_buffer_size);

 if (output_buffer) {
 memset(output_buffer, 0, output_buffer_size); // Zero padding to avoid random noise
 int converted_samples = swr_convert(swr_ctx, &output_buffer, output_samples,
 (const uint8_t**)frame->extended_data, frame->nb_samples);

 if (converted_samples >= 0) {
 output.insert(output.end(), output_buffer, output_buffer + output_buffer_size);
 }
 else {
 std::cerr << "Error: Failed to convert audio samples." << std::endl;
 }
 // Make sure output_buffer is valid before freeing
 if (output_buffer != nullptr) {
 av_free(output_buffer);
 output_buffer = nullptr; // Prevent double-free
 }
 }
 else {
 std::cerr << "Error: Could not allocate output buffer." << std::endl;
 }
 }
 }
 else {
 std::cerr << "Error: Failed to send packet to codec context." << std::endl;
 }
 }
 av_packet_unref(&packet);
 }

 int frame_width = av_get_bytes_per_sample(AV_SAMPLE_FMT_S16) * 2; // Use 2 bytes per sample and 2 channels

 std::map metadata = {
 {"sample_width", 2}, // S16 format has 2 bytes per sample
 {"frame_rate", codec_ctx->sample_rate}, // Use the input sample rate
 {"channels", 2}, // Assuming stereo output
 {"frame_width", frame_width}
 };

 av_frame_free(&frame);
 swr_free(&swr_ctx);
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);

 return AudioSegment(static_cast<const>(output.data()), output.size(), metadata);
}









std::ofstream AudioSegment::export_segment(const std::string& out_f,
 const std::string& format,
 const std::string& codec,
 const std::string& bitrate,
 const std::vector& parameters,
 const std::map& tags,
 const std::string& id3v2_version,
 const std::string& cover) {

 av_log_set_level(AV_LOG_DEBUG);
 AVCodecContext* codec_ctx = nullptr;
 AVFormatContext* format_ctx = nullptr;
 AVStream* stream = nullptr;
 AVFrame* frame = nullptr;
 AVPacket* pkt = nullptr;
 SwrContext* swr_ctx = nullptr;
 int ret;

 // Initialize format context
 if (avformat_alloc_output_context2(&format_ctx, nullptr, format.c_str(), out_f.c_str()) < 0) {
 throw std::runtime_error("Could not allocate format context.");
 }

 // Find encoder
 const AVCodec* codec_ptr = avcodec_find_encoder_by_name(codec.c_str());
 if (!codec_ptr) {
 throw std::runtime_error("Codec not found.");
 }

 // Add stream
 stream = avformat_new_stream(format_ctx, codec_ptr);
 if (!stream) {
 throw std::runtime_error("Failed to create new stream.");
 }

 // Allocate codec context
 codec_ctx = avcodec_alloc_context3(codec_ptr);
 if (!codec_ctx) {
 throw std::runtime_error("Could not allocate audio codec context.");
 }

 // Set codec parameters
 codec_ctx->bit_rate = std::stoi(bitrate);
 codec_ctx->sample_rate = this->get_frame_rate(); // Ensure this returns the correct sample rate
 av_channel_layout_default(&codec_ctx->ch_layout, 2);
 codec_ctx->sample_fmt = codec_ptr->sample_fmts ? codec_ptr->sample_fmts[0] : AV_SAMPLE_FMT_FLTP;

 // Open codec
 if (avcodec_open2(codec_ctx, codec_ptr, nullptr) < 0) {
 throw std::runtime_error("Could not open codec.");
 }

 // Set codec parameters to the stream
 if (avcodec_parameters_from_context(stream->codecpar, codec_ctx) < 0) {
 throw std::runtime_error("Could not initialize stream codec parameters.");
 }

 // Open output file
 std::ofstream out_file(out_f, std::ios::binary);
 if (!out_file) {
 throw std::runtime_error("Failed to open output file.");
 }

 if (!(format_ctx->oformat->flags & AVFMT_NOFILE)) {
 if (avio_open(&format_ctx->pb, out_f.c_str(), AVIO_FLAG_WRITE) < 0) {
 throw std::runtime_error("Could not open output file.");
 }
 }

 // Write file header
 if (avformat_write_header(format_ctx, nullptr) < 0) {
 throw std::runtime_error("Error occurred when opening output file.");
 }

 // Initialize packet
 pkt = av_packet_alloc();
 if (!pkt) {
 throw std::runtime_error("Could not allocate AVPacket.");
 }

 // Initialize frame
 frame = av_frame_alloc();
 if (!frame) {
 throw std::runtime_error("Could not allocate AVFrame.");
 }
 frame->nb_samples = codec_ctx->frame_size;
 frame->format = codec_ctx->sample_fmt;
 frame->ch_layout = codec_ctx->ch_layout;

 // Allocate data buffer
 if (av_frame_get_buffer(frame, 0) < 0) {
 throw std::runtime_error("Could not allocate audio data buffers.");
 }

 // Initialize SwrContext for resampling
 swr_ctx = swr_alloc();
 if (!swr_ctx) {
 throw std::runtime_error("Could not allocate SwrContext.");
 }

 // Set options for resampling
 av_opt_set_chlayout(swr_ctx, "in_chlayout", &codec_ctx->ch_layout, 0);
 av_opt_set_chlayout(swr_ctx, "out_chlayout", &codec_ctx->ch_layout, 0);
 av_opt_set_int(swr_ctx, "in_sample_rate", codec_ctx->sample_rate, 0);
 av_opt_set_int(swr_ctx, "out_sample_rate", codec_ctx->sample_rate, 0);
 av_opt_set_sample_fmt(swr_ctx, "in_sample_fmt", AV_SAMPLE_FMT_S16, 0); // Assuming input is S16
 av_opt_set_sample_fmt(swr_ctx, "out_sample_fmt", codec_ctx->sample_fmt, 0);

 // Initialize the resampling context
 if (swr_init(swr_ctx) < 0) {
 throw std::runtime_error("Failed to initialize SwrContext.");
 }

 int samples_read = 0;
 int total_samples = data_.size() / (av_get_bytes_per_sample(AV_SAMPLE_FMT_S16) * 2); // Assuming input is stereo

 while (samples_read < total_samples) {
 if (av_frame_make_writable(frame) < 0) {
 throw std::runtime_error("Frame not writable.");
 }

 int num_samples = std::min(codec_ctx->frame_size, total_samples - samples_read);

 // Prepare input data
 const uint8_t* input_data[2] = { reinterpret_cast<const>(data_.data() + samples_read * av_get_bytes_per_sample(AV_SAMPLE_FMT_S16) * 2), nullptr };
 int output_samples = swr_convert(swr_ctx, frame->data, frame->nb_samples,
 input_data, num_samples);

 if (output_samples < 0) {
 throw std::runtime_error("Error converting audio samples.");
 }

 frame->nb_samples = output_samples;

 // Send the frame for encoding
 if (avcodec_send_frame(codec_ctx, frame) < 0) {
 throw std::runtime_error("Error sending frame for encoding.");
 }

 // Receive and write packets
 while (avcodec_receive_packet(codec_ctx, pkt) >= 0) {
 out_file.write(reinterpret_cast(pkt->data), pkt->size);
 av_packet_unref(pkt);
 }

 samples_read += num_samples;
 }

 // Flush the encoder
 if (avcodec_send_frame(codec_ctx, nullptr) < 0) {
 throw std::runtime_error("Error flushing the encoder.");
 }

 while (avcodec_receive_packet(codec_ctx, pkt) >= 0) {
 out_file.write(reinterpret_cast(pkt->data), pkt->size);
 av_packet_unref(pkt);
 }

 // Write file trailer
 av_write_trailer(format_ctx);

 // Cleanup
 av_frame_free(&frame);
 av_packet_free(&pkt);
 swr_free(&swr_ctx);
 avcodec_free_context(&codec_ctx);

 if (!(format_ctx->oformat->flags & AVFMT_NOFILE)) {
 avio_closep(&format_ctx->pb);
 }
 avformat_free_context(format_ctx);

 out_file.close();
 return out_file;
}

//declaration
/*
std::ofstream export_segment(const std::string& out_f,
 const std::string& format = "mp3",
 const std::string& codec = "libmp3lame",
 const std::string& bitrate = "128000",
 const std::vector& parameters = {},
 const std::map& tags = {},
 const std::string& id3v2_version = "4",
 const std::string& cover = "");
*/
</const></const></char>

This code only works for mp3 format. I also want to export to aac,ogg,flv,wav and any other popular formats.


-
Java uses FFmpegRecoder to encode frames into H264 streams
5 septembre 2024, par zhang1973I want to obtain the Frame from the video stream, process it, use FFmpegRecoder to encode it into an H264 stream, and transmit it to the front-end. But I found that the AVPacket obtained directly using grabber.grabAVPacket can be converted into H264 stream and played normally. The H264 stream encoded using FFmpegRecoder cannot be played.


Here is my Code :


private FFmpegFrameRecorder recorder;
 private ByteArrayOutputStream outputStream = new ByteArrayOutputStream();;
 private boolean createRecoder(Frame frame){
 recorder = new FFmpegFrameRecorder(outputStream, frame.imageWidth, frame.imageHeight);
 recorder.setVideoCodec(avcodec.AV_CODEC_ID_H264);
 recorder.setFormat("h264"); //"h264"); //
 recorder.setFrameRate(30);
 recorder.setPixelFormat(avutil.AV_PIX_FMT_YUV420P);
 recorder.setVideoBitrate(4000 * 1000); // 设置比特率为4000 kbps
 recorder.setVideoOption("preset", "ultrafast"); // 设置编码器预设，"ultrafast"是最快的，"veryslow"是最慢但质量最好
 recorder.setAudioChannels(0);

 try {
 recorder.start();
 return recorderStatus = true;
 } catch (org.bytedeco.javacv.FrameRecorder.Exception e1) {
 log.info("启动转码录制器失败", e1);
 MediaService.cameras.remove(cameraDto.getMediaKey());
 e1.printStackTrace();
 }

 return recorderStatus = false;
 }

 private boolean slow = false;
 protected void transferStream2H264() throws FFmpegFrameGrabber.Exception {

 // 初始化和拉去图像的方法
 log.info(" create grabber ");
 if (!createGrabber()) {
 log.error(" == > ");
 return;
 }
 transferFlag = true;

 if(!createRecoder(grabber.grab())){
 return;
 }

 try {
 grabber.flush();
 } catch (Exception e) {
 log.info("清空拉流器缓存失败", e);
 e.printStackTrace();
 }

 if (header == null) {
 header = bos.toByteArray();
 slow = true;
// System.out.println("Header1");
// System.out.println(header);
 bos.reset();
 }else{
 System.out.println("Header2");
 System.out.println(header);
 }

 running = true;

 // 事实更新所有的连接数
 listenClient();

 long startTime = 0;
 long videoTS = 0;

 for (; running && grabberStatus;) {
 try {
 if (transferFlag) {
 long startGrab = System.currentTimeMillis();
 //视频采集器
// AVPacket pkt = grabber.grabPacket();
 Frame frame = grabber.grab();
 recorder.record(frame);
 byte[] videoData = outputStream.toByteArray();
 if ((System.currentTimeMillis() - startGrab) > 5000) {
 log.info("\r\n{}\r\n视频流网络异常>>>", cameraDto.getUrl());
 closeMedia();
 break;
 }

 videoTS = 1000 * (System.currentTimeMillis() - startTime);


 if (startTime == 0) {
 startTime = System.currentTimeMillis();
 }
 videoTS = 1000 * (System.currentTimeMillis() - startTime);

 byte[] rbuffer = videoData;
 readSize = videoData.length;

 if(spsdata == null || ppsdata == null){
 movePos = 0;
 lastPos = 0;
 isNewPack = true;
 while(movePos < readSize){
 if (rbuffer[movePos] == 0 && rbuffer[movePos + 1] == 0 && rbuffer[movePos + 2] == 1) {
 findCode = true;
 skipLen = 3;
 mCurFrameFirstByte = (int)(0xff & rbuffer[movePos + skipLen]);
 } else if (rbuffer[movePos] == 0 && rbuffer[movePos + 1] == 0 && rbuffer[movePos + 2] == 0 && rbuffer[movePos + 3] == 1) {
 findCode = true;
 skipLen = 4;
 mCurFrameFirstByte = (int)(0xff & rbuffer[movePos + skipLen]);
 } else {
 skipLen = 1;
 }

 if(!isFirstFind && isNewPack && findCode){
 mFrameFirstByte = mCurFrameFirstByte;
 findCode = false;
 isNewPack = false;
 mNaluType = mFrameFirstByte & 0x1f;
 if(mNaluType != MediaConstant.NALU_TYPE_SEI &&
 mNaluType != MediaConstant.NALU_TYPE_SPS &&
 mNaluType != MediaConstant.NALU_TYPE_PPS &&
 mNaluType != MediaConstant.NALU_TYPE_IDR){
 startCounter++;
 break;
 }
 }

 if(isFirstFind){
 isFirstFind = false;
 findCode = false;
 mFrameFirstByte = mCurFrameFirstByte;
 }

 if(findCode){
 startCounter++;
 mNaluType = mFrameFirstByte & 0x1f;

 findCode = false;
 mFrameLen = (movePos - lastPos);
 if(mNaluType == MediaConstant.NALU_TYPE_IDR){
 mFrameLen = readSize - movePos;
 }

 if(mNaluType != MediaConstant.NALU_TYPE_SEI &&
 mNaluType != MediaConstant.NALU_TYPE_SPS &&
 mNaluType != MediaConstant.NALU_TYPE_PPS &&
 mNaluType != MediaConstant.NALU_TYPE_IDR){
 System.out.println(" one packe many frames ---> type: " + mNaluType + " jump out ");
 break;
 }
 if(mNaluType == MediaConstant.NALU_TYPE_SPS){
 if(null == spsdata){
 spsdata = new byte[mFrameLen];
 System.arraycopy(rbuffer, lastPos, spsdata, 0, mFrameLen);
 }
 }
 if(mNaluType == MediaConstant.NALU_TYPE_PPS){

 if(null == ppsdata){
 ppsdata = new byte[mFrameLen];
 System.arraycopy(rbuffer, lastPos, ppsdata, 0, mFrameLen);
 }
 }

 lastPos = movePos;
 mFrameFirstByte = mCurFrameFirstByte;
 mNaluType = mFrameFirstByte & 0x1f;
 if(mNaluType == MediaConstant.NALU_TYPE_IDR){
 mFrameLen = readSize - movePos;
 startCounter++;

 break;
 }
 }

 movePos += skipLen;
 isNewPack = false;
 }
 }

 sendFrameData(rbuffer);
// }
// av_packet_unref(pkt);
// }

// }
 } else {
 }
 } catch (Exception e) {
 grabberStatus = false;
 MediaService.cameras.remove(cameraDto.getMediaKey());
 } catch (FFmpegFrameRecorder.Exception e) {
 throw new RuntimeException(e);
 }
 }

 try {
 grabber.close();
 bos.close();
 } catch (org.bytedeco.javacv.FrameRecorder.Exception e) {
 e.printStackTrace();
 } catch (Exception e) {
 e.printStackTrace();
 } catch (IOException e) {
 e.printStackTrace();
 } finally {
 closeMedia();
 }
 log.info("关闭媒体流-javacv，{} ", cameraDto.getUrl());
 }


-
C++ ffmpeg - export to wav error : Invalid PCM packet, data has size 2 but at least a size of 4 was expected
9 septembre 2024, par Chris PC++ code :


AudioSegment AudioSegment::from_file(const std::string& file_path, const std::string& format, const std::string& codec,
 const std::map& parameters, int start_second, int duration) {

 avformat_network_init();
 av_log_set_level(AV_LOG_ERROR); // Adjust logging level as needed

 AVFormatContext* format_ctx = nullptr;
 if (avformat_open_input(&format_ctx, file_path.c_str(), nullptr, nullptr) != 0) {
 std::cerr << "Error: Could not open audio file." << std::endl;
 return AudioSegment(); // Return an empty AudioSegment on failure
 }

 if (avformat_find_stream_info(format_ctx, nullptr) < 0) {
 std::cerr << "Error: Could not find stream information." << std::endl;
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 int audio_stream_index = -1;
 for (unsigned int i = 0; i < format_ctx->nb_streams; i++) {
 if (format_ctx->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
 audio_stream_index = i;
 break;
 }
 }

 if (audio_stream_index == -1) {
 std::cerr << "Error: Could not find audio stream." << std::endl;
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 AVCodecParameters* codec_par = format_ctx->streams[audio_stream_index]->codecpar;
 const AVCodec* my_codec = avcodec_find_decoder(codec_par->codec_id);
 AVCodecContext* codec_ctx = avcodec_alloc_context3(my_codec);

 if (!codec_ctx) {
 std::cerr << "Error: Could not allocate codec context." << std::endl;
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 if (avcodec_parameters_to_context(codec_ctx, codec_par) < 0) {
 std::cerr << "Error: Could not initialize codec context." << std::endl;
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 if (avcodec_open2(codec_ctx, my_codec, nullptr) < 0) {
 std::cerr << "Error: Could not open codec." << std::endl;
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 SwrContext* swr_ctx = swr_alloc();
 if (!swr_ctx) {
 std::cerr << "Error: Could not allocate SwrContext." << std::endl;
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }
 codec_ctx->sample_rate = 44100;
 // Set up resampling context to convert to S16 format with 2 bytes per sample
 av_opt_set_chlayout(swr_ctx, "in_chlayout", &codec_ctx->ch_layout, 0);
 av_opt_set_int(swr_ctx, "in_sample_rate", codec_ctx->sample_rate, 0);
 av_opt_set_sample_fmt(swr_ctx, "in_sample_fmt", codec_ctx->sample_fmt, 0);

 AVChannelLayout dst_ch_layout;
 av_channel_layout_copy(&dst_ch_layout, &codec_ctx->ch_layout);
 av_channel_layout_uninit(&dst_ch_layout);
 av_channel_layout_default(&dst_ch_layout, 2);

 av_opt_set_chlayout(swr_ctx, "out_chlayout", &dst_ch_layout, 0);
 av_opt_set_int(swr_ctx, "out_sample_rate", codec_ctx->sample_rate, 0); // Match input sample rate
 av_opt_set_sample_fmt(swr_ctx, "out_sample_fmt", AV_SAMPLE_FMT_S16, 0); // Force S16 format

 if (swr_init(swr_ctx) < 0) {
 std::cerr << "Error: Failed to initialize the resampling context" << std::endl;
 swr_free(&swr_ctx);
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 AVPacket packet;
 AVFrame* frame = av_frame_alloc();
 if (!frame) {
 std::cerr << "Error: Could not allocate frame." << std::endl;
 swr_free(&swr_ctx);
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);
 return AudioSegment();
 }

 std::vector<char> output;
 while (av_read_frame(format_ctx, &packet) >= 0) {
 if (packet.stream_index == audio_stream_index) {
 if (avcodec_send_packet(codec_ctx, &packet) == 0) {
 while (avcodec_receive_frame(codec_ctx, frame) == 0) {
 if (frame->pts != AV_NOPTS_VALUE) {
 frame->pts = av_rescale_q(frame->pts, codec_ctx->time_base, format_ctx->streams[audio_stream_index]->time_base);
 }

 uint8_t* output_buffer;
 int output_samples = av_rescale_rnd(
 swr_get_delay(swr_ctx, codec_ctx->sample_rate) + frame->nb_samples,
 codec_ctx->sample_rate, codec_ctx->sample_rate, AV_ROUND_UP);

 int output_buffer_size = av_samples_get_buffer_size(
 nullptr, 2, output_samples, AV_SAMPLE_FMT_S16, 1);

 output_buffer = (uint8_t*)av_malloc(output_buffer_size);

 if (output_buffer) {
 memset(output_buffer, 0, output_buffer_size); // Zero padding to avoid random noise
 int converted_samples = swr_convert(swr_ctx, &output_buffer, output_samples,
 (const uint8_t**)frame->extended_data, frame->nb_samples);

 if (converted_samples >= 0) {
 output.insert(output.end(), output_buffer, output_buffer + output_buffer_size);
 }
 else {
 std::cerr << "Error: Failed to convert audio samples." << std::endl;
 }
 // Make sure output_buffer is valid before freeing
 if (output_buffer != nullptr) {
 av_free(output_buffer);
 output_buffer = nullptr; // Prevent double-free
 }
 }
 else {
 std::cerr << "Error: Could not allocate output buffer." << std::endl;
 }
 }
 }
 else {
 std::cerr << "Error: Failed to send packet to codec context." << std::endl;
 }
 }
 av_packet_unref(&packet);
 }

 int frame_width = av_get_bytes_per_sample(AV_SAMPLE_FMT_S16) * 2; // Use 2 bytes per sample and 2 channels

 std::map metadata = {
 {"sample_width", 2}, // S16 format has 2 bytes per sample
 {"frame_rate", codec_ctx->sample_rate}, // Use the input sample rate
 {"channels", 2}, // Assuming stereo output
 {"frame_width", frame_width}
 };

 av_frame_free(&frame);
 swr_free(&swr_ctx);
 avcodec_free_context(&codec_ctx);
 avformat_close_input(&format_ctx);

 return AudioSegment(static_cast<const>(output.data()), output.size(), metadata);
}

std::ofstream AudioSegment::export_segment_to_wav_file(const std::string& out_f) {
 std::cout << this->get_channels() << std::endl;
 av_log_set_level(AV_LOG_ERROR);
 AVCodecContext* codec_ctx = nullptr;
 AVFormatContext* format_ctx = nullptr;
 AVStream* stream = nullptr;
 AVFrame* frame = nullptr;
 AVPacket* pkt = nullptr;
 int ret;

 // Initialize format context for WAV
 if (avformat_alloc_output_context2(&format_ctx, nullptr, "wav", out_f.c_str()) < 0) {
 throw std::runtime_error("Could not allocate format context.");
 }

 // Find encoder for PCM
 const AVCodec* codec_ptr = avcodec_find_encoder(AV_CODEC_ID_PCM_S16LE);
 if (!codec_ptr) {
 throw std::runtime_error("PCM encoder not found.");
 }

 // Add stream
 stream = avformat_new_stream(format_ctx, codec_ptr);
 if (!stream) {
 throw std::runtime_error("Failed to create new stream.");
 }

 // Allocate codec context
 codec_ctx = avcodec_alloc_context3(codec_ptr);
 if (!codec_ctx) {
 throw std::runtime_error("Could not allocate audio codec context.");
 }

 // Set codec parameters for PCM
 codec_ctx->bit_rate = 128000; // Bitrate
 codec_ctx->sample_rate = this->get_frame_rate(); // Use correct sample rate
 codec_ctx->ch_layout.nb_channels = this->get_channels(); // Set the correct channel count

 // Set the channel layout: stereo or mono
 if (this->get_channels() == 2) {
 av_channel_layout_default(&codec_ctx->ch_layout, 2); // Stereo layout
 }
 else {
 av_channel_layout_default(&codec_ctx->ch_layout, 1); // Mono layout
 }

 codec_ctx->sample_fmt = AV_SAMPLE_FMT_S16; // PCM 16-bit format

 // Open codec
 if (avcodec_open2(codec_ctx, codec_ptr, nullptr) < 0) {
 throw std::runtime_error("Could not open codec.");
 }

 // Set codec parameters to the stream
 if (avcodec_parameters_from_context(stream->codecpar, codec_ctx) < 0) {
 throw std::runtime_error("Could not initialize stream codec parameters.");
 }

 // Open output file
 std::ofstream out_file(out_f, std::ios::binary);
 if (!out_file) {
 throw std::runtime_error("Failed to open output file.");
 }

 if (!(format_ctx->oformat->flags & AVFMT_NOFILE)) {
 if (avio_open(&format_ctx->pb, out_f.c_str(), AVIO_FLAG_WRITE) < 0) {
 throw std::runtime_error("Could not open output file.");
 }
 }

 // Write file header
 if (avformat_write_header(format_ctx, nullptr) < 0) {
 throw std::runtime_error("Error occurred when writing file header.");
 }

 // Initialize packet
 pkt = av_packet_alloc();
 if (!pkt) {
 throw std::runtime_error("Could not allocate AVPacket.");
 }

 // Initialize frame
 frame = av_frame_alloc();
 if (!frame) {
 throw std::runtime_error("Could not allocate AVFrame.");
 }

 // Set the frame properties
 frame->format = codec_ctx->sample_fmt;
 frame->ch_layout = codec_ctx->ch_layout;

 // Number of audio samples available in the data buffer
 int total_samples = data_.size() / (av_get_bytes_per_sample(AV_SAMPLE_FMT_S16) * codec_ctx->ch_layout.nb_channels);
 int samples_read = 0;

 // Set the number of samples per frame dynamically based on the input data
 while (samples_read < total_samples) {
 // Determine how many samples to read in this iteration (don't exceed the total sample count)
 int num_samples = std::min(codec_ctx->frame_size, total_samples - samples_read);
 if (num_samples == 0) {
 num_samples = 1024;
 codec_ctx->frame_size = 1024;
 }
 // Ensure num_samples is not zero
 if (num_samples <= 0) {
 throw std::runtime_error("Invalid number of samples in frame.");
 }

 // Set the number of samples in the frame
 frame->nb_samples = num_samples;

 // Allocate the frame buffer based on the number of samples
 ret = av_frame_get_buffer(frame, 0);
 if (ret < 0) {
 std::cerr << "Error allocating frame buffer: " << ret << std::endl;
 throw std::runtime_error("Could not allocate audio data buffers.");
 }

 // Copy the audio data into the frame's buffer (interleaving if necessary)
 /*if (codec_ctx->ch_layout.nb_channels == 2) {
 // If stereo, interleave planar data into packed format
 for (int i = 0; i < num_samples; ++i) {
 ((int16_t*)frame->data[0])[2 * i] = ((int16_t*)data_.data())[i]; // Left channel
 ((int16_t*)frame->data[0])[2 * i + 1] = ((int16_t*)data_.data())[total_samples + i]; // Right channel
 }
 }
 else {
 // For mono or packed data, directly copy the samples
 std::memcpy(frame->data[0], data_.data() + samples_read * av_get_bytes_per_sample(AV_SAMPLE_FMT_S16) * codec_ctx->ch_layout.nb_channels,
 num_samples * av_get_bytes_per_sample(AV_SAMPLE_FMT_S16) * codec_ctx->ch_layout.nb_channels);
 }
 */
 std::memcpy(frame->data[0], data_.data() + samples_read * av_get_bytes_per_sample(AV_SAMPLE_FMT_S16) * codec_ctx->ch_layout.nb_channels,
 num_samples * av_get_bytes_per_sample(AV_SAMPLE_FMT_S16) * codec_ctx->ch_layout.nb_channels);

 // Send the frame for encoding
 ret = avcodec_send_frame(codec_ctx, frame);
 if (ret < 0) {
 std::cerr << "Error sending frame for encoding: " << ret << std::endl;
 throw std::runtime_error("Error sending frame for encoding.");
 }

 // Receive and write encoded packets
 while (ret >= 0) {
 ret = avcodec_receive_packet(codec_ctx, pkt);
 if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
 break;
 }
 else if (ret < 0) {
 throw std::runtime_error("Error during encoding.");
 }

 out_file.write(reinterpret_cast(pkt->data), pkt->size);
 av_packet_unref(pkt);
 }

 samples_read += num_samples;
 }

 // Flush the encoder
 if (avcodec_send_frame(codec_ctx, nullptr) < 0) {
 throw std::runtime_error("Error flushing the encoder.");
 }

 while (avcodec_receive_packet(codec_ctx, pkt) >= 0) {
 out_file.write(reinterpret_cast(pkt->data), pkt->size);
 av_packet_unref(pkt);
 }

 // Write file trailer
 av_write_trailer(format_ctx);

 // Cleanup
 av_frame_free(&frame);
 av_packet_free(&pkt);
 avcodec_free_context(&codec_ctx);

 if (!(format_ctx->oformat->flags & AVFMT_NOFILE)) {
 avio_closep(&format_ctx->pb);
 }
 avformat_free_context(format_ctx);

 out_file.close();
 return out_file;
}

</const></char>

Run code :


#include "audio_segment.h"
#include "effects.h"
#include "playback.h"
#include "cppaudioop.h"
#include "exceptions.h"
#include "generators.h"
#include "silence.h"
#include "utils.h"

#include <iostream>
#include <filesystem>

using namespace cppdub;

int main() {
 try {
 // Load the source audio file
 AudioSegment seg_1 = AudioSegment::from_file("../data/test10.mp3");
 std::string out_file_name = "ah-ah-ah.wav";

 // Export the audio segment to a new file with specified settings
 //seg_1.export_segment(out_file_name, "mp3");
 seg_1.export_segment_to_wav_file(out_file_name);


 // Optionally play the audio segment to verify
 // play(seg_1);

 // Load the exported audio file
 AudioSegment seg_2 = AudioSegment::from_file(out_file_name);

 // Play segments
 //play(seg_1);
 play(seg_2);
 }
 catch (const std::exception& e) {
 std::cerr << "An error occurred: " << e.what() << std::endl;
 }

 return 0;
}
</filesystem></iostream>

Error in second call of from_file function :


[pcm_s16le @ 000002d82ca5bfc0] Invalid PCM packet, data has size 2 but at least a size of 4 was expected


The process continue, i call hear the seg_2 with play(seg_2) call, but i can't directly play seg_2 export wav file (from windows explorer).


I had a guess that error may be because packed vs plannar formats missmatch but i am not quit sure. Maybe a swr_convert is necessary.


Syndiquer tout le site
