aacdec: add support for dual mono in Japanese DTV

Japanese DTV uses some non standard extensions in AAC audio.
One example is 'dual mono', which combines two independent
audio into one stereo stream, storing them in left and right channels
respectively.  Historically, dual mono audio has been used for
multi-lingual audio, one for local/native language, and another for english,
and usually the "main" (local language) channel should be output without
any user interactions.

The frames of those dual mono audio are allowed to set
ADTS channel_config field to 0, and just contain two SCE's *WITHOUT* PCE,
which is a non standard extension by Japanese DTV standard.
(ref. ARIB STD-B32 PartII 5.2.3)

This patch adds an AVPacket side data, AV_PKT_DATA_JP_DUALMONO,
which indicates that the AVPacket is likely to contain an audio frame
with the above dual mono extension, and has the parameter to specify
the desired channel selection in that case.
It also makes aacdec to detect dual mono and output just the desired
channel when this side data is attached.
Signed-off-by: Akihiro Tsukada <atsukada@users.sourceforge.net>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>

libavcodec/aac.h

@@ -304,6 +304,15 @@ typedef struct {
     float *output_data[MAX_CHANNELS];                 ///< Points to each element's 'ret' buffer (PCM output).
     /** @} */
 
+
+    /**
+     * @name Japanese DTV specific extension
+     * @{
+     */
+    int enable_jp_dmono; ///< enable japanese DTV specific 'dual mono'
+    int dmono_mode;      ///< select the channel to decode in dual mono.
+    /** @} */
+
     DECLARE_ALIGNED(32, float, temp)[128];
 
     OutputConfiguration oc[2];

libavcodec/aacdec.c

@@ -2375,6 +2375,21 @@ static int parse_adts_frame_header(AACContext *ac, GetBitContext *gb)
                 return -7;
         } else {
             ac->oc[1].m4ac.chan_config = 0;
+            /**
+             * dual mono frames in Japanese DTV can have chan_config 0
+             * WITHOUT specifying PCE.
+             *  thus, set dual mono as default.
+             */
+            if (ac->enable_jp_dmono && ac->oc[0].status == OC_NONE) {
+                layout_map_tags = 2;
+                layout_map[0][0] = layout_map[1][0] = TYPE_SCE;
+                layout_map[0][2] = layout_map[1][2] = AAC_CHANNEL_FRONT;
+                layout_map[0][1] = 0;
+                layout_map[1][1] = 1;
+                if (output_configure(ac, layout_map, layout_map_tags,
+                                     0, OC_TRIAL_FRAME))
+                    return -7;
+            }
         }
         ac->oc[1].m4ac.sample_rate     = hdr_info.sample_rate;
         ac->oc[1].m4ac.sampling_index  = hdr_info.sampling_index;
@@ -2399,6 +2414,8 @@ static int aac_decode_frame_int(AVCodecContext *avctx, void *data,
     enum RawDataBlockType elem_type, elem_type_prev = TYPE_END;
     int err, elem_id;
     int samples = 0, multiplier, audio_found = 0, pce_found = 0;
+    int is_dmono, sce_count = 0;
+    float *tmp = NULL;
 
     if (show_bits(gb, 12) == 0xfff) {
         if (parse_adts_frame_header(ac, gb) < 0) {
@@ -2433,6 +2450,7 @@ static int aac_decode_frame_int(AVCodecContext *avctx, void *data,
         case TYPE_SCE:
             err = decode_ics(ac, &che->ch[0], gb, 0, 0);
             audio_found = 1;
+            sce_count++;
             break;
 
         case TYPE_CPE:
@@ -2511,6 +2529,20 @@ static int aac_decode_frame_int(AVCodecContext *avctx, void *data,
     multiplier = (ac->oc[1].m4ac.sbr == 1) ? ac->oc[1].m4ac.ext_sample_rate > ac->oc[1].m4ac.sample_rate : 0;
     samples <<= multiplier;
 
+    /* for dual-mono audio (SCE + SCE) */
+    is_dmono = ac->enable_jp_dmono && sce_count == 2 &&
+               ac->oc[1].channel_layout == (AV_CH_FRONT_LEFT | AV_CH_FRONT_RIGHT);
+
+    if (is_dmono) {
+        if (ac->dmono_mode == 0) {
+            tmp = ac->output_data[1];
+            ac->output_data[1] = ac->output_data[0];
+        } else if (ac->dmono_mode == 1) {
+            tmp = ac->output_data[0];
+            ac->output_data[0] = ac->output_data[1];
+        }
+    }
+
     if (samples) {
         /* get output buffer */
         ac->frame.nb_samples = samples;
@@ -2533,6 +2565,13 @@ static int aac_decode_frame_int(AVCodecContext *avctx, void *data,
     }
     *got_frame_ptr = !!samples;
 
+    if (is_dmono) {
+        if (ac->dmono_mode == 0)
+            ac->output_data[1] = tmp;
+        else if (ac->dmono_mode == 1)
+            ac->output_data[0] = tmp;
+    }
+
     if (ac->oc[1].status && audio_found) {
         avctx->sample_rate = ac->oc[1].m4ac.sample_rate << multiplier;
         avctx->frame_size = samples;
@@ -2565,6 +2604,10 @@ static int aac_decode_frame(AVCodecContext *avctx, void *data,
     const uint8_t *new_extradata = av_packet_get_side_data(avpkt,
                                        AV_PKT_DATA_NEW_EXTRADATA,
                                        &new_extradata_size);
+    int jp_dualmono_size;
+    const uint8_t *jp_dualmono   = av_packet_get_side_data(avpkt,
+                                       AV_PKT_DATA_JP_DUALMONO,
+                                       &jp_dualmono_size);
 
     if (new_extradata && 0) {
         av_free(avctx->extradata);
@@ -2583,6 +2626,11 @@ static int aac_decode_frame(AVCodecContext *avctx, void *data,
         }
     }
 
+    ac->enable_jp_dmono = !!jp_dualmono;
+    ac->dmono_mode = 0;
+    if (jp_dualmono && jp_dualmono_size > 0)
+        ac->dmono_mode = *jp_dualmono;
+
     init_get_bits(&gb, buf, buf_size * 8);
 
     if ((err = aac_decode_frame_int(avctx, data, got_frame_ptr, &gb, avpkt)) < 0)

libavcodec/avcodec.h

@@ -927,6 +927,16 @@ enum AVPacketSideDataType {
      * @endcode
      */
     AV_PKT_DATA_SKIP_SAMPLES=70,
+
+    /**
+     * An AV_PKT_DATA_JP_DUALMONO side data packet indicates that
+     * the packet may contain "dual mono" audio specific to Japanese DTV
+     * and if it is true, recommends only the selected channel to be used.
+     * @code
+     * u8    selected channels (0=mail/left, 1=sub/right, 2=both)
+     * @endcode
+     */
+    AV_PKT_DATA_JP_DUALMONO,
 };
 
 typedef struct AVPacket {

libavcodec/version.h

@@ -27,7 +27,7 @@
  */
 
 #define LIBAVCODEC_VERSION_MAJOR 54
-#define LIBAVCODEC_VERSION_MINOR 55
+#define LIBAVCODEC_VERSION_MINOR 56
 #define LIBAVCODEC_VERSION_MICRO 100
 
 #define LIBAVCODEC_VERSION_INT  AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \