调用科大讯飞API实现文本转语音源码

#include "stdafx.h" #ifdef __cplusplus extern "C" { #endif #include "libavutil/avassert.h" #include "libavutil/audio_fifo.h" #include "libavutil/avstring.h" #include "libavutil/frame.h" #include "libavutil/opt.h" #include "libavutil/avassert.h" #include "libswresample/swresample.h" #include"logger.h" #include "transcode_audio.h" #ifdef __cplusplus }; #endif /** The output bit rate in kbit/s */ #define OUTPUT_BIT_RATE 25000 /** The number of output channels */ #define OUTPUT_CHANNELS 1 static int encode_audio_frame_flush(AVFrame *frame, AVFormatContext *output_format_context, AVCodecContext *output_codec_context, int *data_present); static int current_percent(AVPacket pkt, AVFormatContext *fmt_ctx); static void transcode_callback(TranscodeCallbackFcn fun, int status, int percent, void *identifier); /** Open an input file and the required decoder. */ static int open_input_file(const char *filename, AVFormatContext **input_format_context, AVCodecContext **input_codec_context) { AVCodecContext *avctx; AVCodec *input_codec; int error; /** Open the input file to read from it. */ if ((error = avformat_open_input(input_format_context, filename, NULL, NULL)) < 0) { fprintf(stderr, "Could not open input file '%s' (error '%s')\n", filename, av_err2str(error)); *input_format_context = NULL; return error; } /** Get information on the input file (number of streams etc.). */ if ((error = avformat_find_stream_info(*input_format_context, NULL)) < 0) { fprintf(stderr, "Could not open find stream info (error '%s')\n", av_err2str(error)); avformat_close_input(input_format_context); return error; } //为什么确保只有一个流 /** Make sure that there is only one stream in the input file. */ if ((*input_format_context)->nb_streams != 1) { fprintf(stderr, "Expected one audio input stream, but found %d\n", (*input_format_context)->nb_streams); avformat_close_input(input_format_context); return AVERROR_EXIT; } /** Find a decoder for the audio stream. */ if (!(input_codec = avcodec_find_decoder((*input_format_context)->streams[0]->codecpar->codec_id))) { fprintf(stderr, "Could not find input codec\n"); avformat_close_input(input_format_context); return AVERROR_EXIT; } /** allocate a new decoding context */ avctx = avcodec_alloc_context3(input_codec); if (!avctx) { fprintf(stderr, "Could not allocate a decoding context\n"); avformat_close_input(input_format_context); return AVERROR(ENOMEM); } /*int64_t i = (*input_format_context)->streams[0]->nb_frames; int64_t i2 = avctx->frame_number;*/ /** initialize the stream parameters with demuxer information */ error = avcodec_parameters_to_context(avctx, (*input_format_context)->streams[0]->codecpar); if (error < 0) { avformat_close_input(input_format_context); avcodec_free_context(&avctx); return error; } /** Open the decoder for the audio stream to use it later. */ if ((error = avcodec_open2(avctx, input_codec, NULL)) < 0) { fprintf(stderr, "Could not open input codec (error '%s')\n", av_err2str(error)); avcodec_free_context(&avctx); avformat_close_input(input_format_context); return error; } /** Save the decoder context for easier access later. */ *input_codec_context = avctx; return 0; } /** * Open an output file and the required encoder. * Also set some basic encoder parameters. * Some of these parameters are based on the input file's parameters. */ static int open_output_file(const char *filename, AVCodecContext *input_codec_context, AVFormatContext **output_format_context, AVCodecContext **output_codec_context) { AVCodecContext *avctx = NULL; AVIOContext *output_io_context = NULL; AVStream *stream = NULL; AVCodec *output_codec = NULL; int error; /** Open the output file to write to it. */ if ((error = avio_open(&output_io_context, filename, AVIO_FLAG_WRITE)) < 0) { fprintf(stderr, "Could not open output file '%s' (error '%s')\n", filename, av_err2str(error)); return error; } /** Create a new format context for the output container format. */ if (!(*output_format_context = avformat_alloc_context())) { fprintf(stderr, "Could not allocate output format context\n"); return AVERROR(ENOMEM); } /** Associate the output file (pointer) with the container format context. */ (*output_format_context)->pb = output_io_context; /** Guess the desired container format based on the file extension. */ if (!((*output_format_context)->oformat = av_guess_format(NULL, filename, NULL))) { fprintf(stderr, "Could not find output file format\n"); goto cleanup; } av_strlcpy((*output_format_context)->filename, filename, sizeof((*output_format_context)->filename)); /** Find the encoder to be used by its name. */ if (!(output_codec = avcodec_find_encoder(AV_CODEC_ID_MP3))) { fprintf(stderr, "Could not find an AAC encoder.\n"); goto cleanup; } /** Create a new audio stream in the output file container. */ if (!(stream = avformat_new_stream(*output_format_context, NULL))) { fprintf(stderr, "Could not create new stream\n"); error = AVERROR(ENOMEM); goto cleanup; } avctx = avcodec_alloc_context3(output_codec); if (!avctx) { fprintf(stderr, "Could not allocate an encoding context\n"); error = AVERROR(ENOMEM); goto cleanup; } /** * Set the basic encoder parameters. * The input file's sample rate is used to avoid a sample rate conversion. */ avctx->channels = OUTPUT_CHANNELS; avctx->channel_layout = av_get_default_channel_layout(OUTPUT_CHANNELS); avctx->sample_rate = input_codec_context->sample_rate; avctx->sample_fmt = output_codec->sample_fmts[0]; avctx->bit_rate = OUTPUT_BIT_RATE; /** Allow the use of the experimental AAC encoder */ avctx->strict_std_compliance = FF_COMPLIANCE_EXPERIMENTAL; /** Set the sample rate for the container. */ stream->time_base.den = input_codec_context->sample_rate; stream->time_base.num = 1; /** * Some container formats (like MP4) require global headers to be present * Mark the encoder so that it behaves accordingly. */ if ((*output_format_context)->oformat->flags & AVFMT_GLOBALHEADER) avctx->flags |= AV_CODEC_FLAG_GLOBAL_HEADER; /** Open the encoder for the audio stream to use it later. */ if ((error = avcodec_open2(avctx, output_codec, NULL)) < 0) { fprintf(stderr, "Could not open output codec (error '%s')\n", av_err2str(error)); goto cleanup; } error = avcodec_parameters_from_context(stream->codecpar, avctx); if (error < 0) { fprintf(stderr, "Could not initialize stream parameters\n"); goto cleanup; } /** Save the encoder context for easier access later. */ *output_codec_context = avctx; return 0; cleanup: avcodec_free_context(&avctx); avio_closep(&(*output_format_context)->pb); avformat_free_context(*output_format_context); *output_format_context = NULL; return error < 0 ? error : AVERROR_EXIT; } /** Initialize one data packet for reading or writing. */ static void init_packet(AVPacket *packet) { av_init_packet(packet); /** Set the packet data and size so that it is recognized as being empty. */ packet->data = NULL; packet->size = 0; } /** Initialize one audio frame for reading from the input file */ static int init_input_frame(AVFrame **frame) { if (!(*frame = av_frame_alloc())) { fprintf(stderr, "Could not allocate input frame\n"); return AVERROR(ENOMEM); } return 0; } /** * Initialize the audio resampler based on the input and output codec settings. * If the input and output sample formats differ, a conversion is required * libswresample takes care of this, but requires initialization. */ static int init_resampler(AVCodecContext *input_codec_context, AVCodecContext *output_codec_context, SwrCont