本文共 6308 字，大约阅读时间需要 21 分钟。

AAC是mp4的音频格式，而安防摄像机基本上使用G711等编码，想要封装成mp4供web预览，就需要跨越g711转AAC的这个难关。

ffmpeg作为音视频界的泰斗，可以帮助我们实现这一功能。

代码流程如下：

重点分析init_resampler/read_decode_convert_and_store/load_encode_and_write 这3个函数为转码的核心函数。

init_resampler主要是进行格式转换使用，AAC现在ffmpeg使用的格式是AV_SAMPLE_FMT_FLTP格式，其他格式好像都不行，而 一般PCM使用AV_SAMPLE_FMT_S16格式，所以需要进行格式转换。

/* Convert the input samples to the desired output sample format.		* This requires a temporary storage provided by converted_input_samples. */		if (convert_samples((const uint8_t**)input_frame->extended_data, converted_input_samples,			input_frame->nb_samples, resampler_context))			goto cleanup;

read_decode_convert_and_store的核心是把读取的音频格式解码成pcm

static int decode_audio_frame(AVFrame *frame,	AVFormatContext *input_format_context,	AVCodecContext *input_codec_context,	int *data_present, int *finished){	/* Packet used for temporary storage. */	AVPacket input_packet;	int error;	init_packet(&input_packet);	/* Read one audio frame from the input file into a temporary packet. */	if ((error = av_read_frame(input_format_context, &input_packet)) < 0) {		/* If we are at the end of the file, flush the decoder below. */		if (error == AVERROR_EOF)			*finished = 1;		else {			fprintf(stderr, "Could not read frame (error '%d')\n",				error);			return error;		}	}	/* Send the audio frame stored in the temporary packet to the decoder.	* The input audio stream decoder is used to do this. */	if ((error = avcodec_send_packet(input_codec_context, &input_packet)) < 0) {		fprintf(stderr, "Could not send packet for decoding (error '%d')\n",			error);		return error;	}	/* Receive one frame from the decoder. */	error = avcodec_receive_frame(input_codec_context, frame);	/* If the decoder asks for more data to be able to decode a frame,	* return indicating that no data is present. */	if (error == AVERROR(EAGAIN)) {		error = 0;		goto cleanup;		/* If the end of the input file is reached, stop decoding. */	}	else if (error == AVERROR_EOF) {		*finished = 1;		error = 0;		goto cleanup;	}	else if (error < 0) {		fprintf(stderr, "Could not decode frame (error '%d')\n",			error);		goto cleanup;		/* Default case: Return decoded data. */	}	else {		*data_present = 1;		goto cleanup;	}cleanup:	av_packet_unref(&input_packet);	return error;}

load_encode_and_write 的核心是把pcm编码成AAC编码

static int encode_audio_frame(AVFrame *frame,	AVFormatContext *output_format_context,	AVCodecContext *output_codec_context,	int *data_present){	/* Packet used for temporary storage. */	AVPacket output_packet;	int error;	init_packet(&output_packet);	/* Set a timestamp based on the sample rate for the container. */	if (frame) {		frame->pts = pts;		pts += frame->nb_samples;	}	/* Send the audio frame stored in the temporary packet to the encoder.	* The output audio stream encoder is used to do this. */	error = avcodec_send_frame(output_codec_context, frame);	/* The encoder signals that it has nothing more to encode. */	if (error == AVERROR_EOF) {		error = 0;		goto cleanup;	}	else if (error < 0) {		fprintf(stderr, "Could not send packet for encoding (error '%d')\n",			error);		return error;	}	/* Receive one encoded frame from the encoder. */	error = avcodec_receive_packet(output_codec_context, &output_packet);	/* If the encoder asks for more data to be able to provide an	* encoded frame, return indicating that no data is present. */	if (error == AVERROR(EAGAIN)) {		error = 0;		goto cleanup;		/* If the last frame has been encoded, stop encoding. */	}	else if (error == AVERROR_EOF) {		error = 0;		goto cleanup;	}	else if (error < 0) {		fprintf(stderr, "Could not encode frame (error '%d')\n",			error);		goto cleanup;		/* Default case: Return encoded data. */	}	else {		*data_present = 1;	}	/* Write one audio frame from the temporary packet to the output file. */	if (*data_present &&		(error = av_write_frame(output_format_context, &output_packet)) < 0) {		fprintf(stderr, "Could not write frame (error '%d')\n",			error);		goto cleanup;	}cleanup:	av_packet_unref(&output_packet);	return error;}

重点API解读：

格式转换，主要是对pcm格式进行2次采样，改变采样率，采样位数， 采样通道

swr_alloc_set_opts：设置格式转换的初始化参数

* Create a resampler context for the conversion.	* Set the conversion parameters.	* Default channel layouts based on the number of channels	* are assumed for simplicity (they are sometimes not detected	* properly by the demuxer and/or decoder).	*/	*resample_context = swr_alloc_set_opts(NULL,		av_get_default_channel_layout(output_codec_context->channels),		output_codec_context->sample_fmt,		output_codec_context->sample_rate,		av_get_default_channel_layout(input_codec_context->channels),		input_codec_context->sample_fmt,		input_codec_context->sample_rate,		0, NULL);

av_samples_alloc:分配音频空间，主要分配转换之后保持的音频

重点参数如下

output_codec_context->channels：音频通道数

frame_size：音频采样数

output_codec_context->sample_fmt：音频采样深度

/* Allocate as many pointers as there are audio channels.	* Each pointer will later point to the audio samples of the corresponding	* channels (although it may be NULL for interleaved formats).	*/	if (!((*converted_input_samples) = (uint8_t * *) calloc(output_codec_context->channels,		sizeof(**converted_input_samples)))) {		fprintf(stderr, "Could not allocate converted input sample pointers\n");		return AVERROR(ENOMEM);	}	/* Allocate memory for the samples of all channels in one consecutive	* block for convenience. */	if ((error = av_samples_alloc(*converted_input_samples, NULL,		output_codec_context->channels,		frame_size,		output_codec_context->sample_fmt, 0)) < 0) {		fprintf(stderr,			"Could not allocate converted input samples (error '%d')\n",			error);		av_freep(&(*converted_input_samples)[0]);		free(*converted_input_samples);		return error;	}

swr_convert ：对音频进行二次采样，并输出结果

重点参数如下

resample_context：swr_alloc_set_opts产生的值

converted_data：av_samples_alloc分配的空间

frame_size：采样次数

input_data：输入的音频

frame_size：输入的采样次数

/* Convert the samples using the resampler. */	if ((error = swr_convert(resample_context,		converted_data, frame_size,		input_data, frame_size)) < 0) {		fprintf(stderr, "Could not convert input samples (error '%d')\n",			error);		return error;	}

转载地址：http://ymvad.baihongyu.com/