hlrc

/*

 * This file is part of hlrc_server

 *

 * Copyright(c) sschulz <AT> techfak.uni-bielefeld.de

 * http://opensource.cit-ec.de/projects/hlrc_server

 *

 * This file may be licensed under the terms of the

 * GNU General Public License Version 3 (the ``GPL''),

 * or (at your option) any later version.

 *

 * Software distributed under the License is distributed

 * on an ``AS IS'' basis, WITHOUT WARRANTY OF ANY KIND, either

 * express or implied. See the GPL for the specific language

 * governing rights and limitations.

 *

 * You should have received a copy of the GPL along with this

 * program. If not, go to http://www.gnu.org/licenses/gpl.html

 * or write to the Free Software Foundation, Inc.,

 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.

 *

 * The development of this software was supported by the

 * Excellence Cluster EXC 277 Cognitive Interaction Technology.

 * The Excellence Cluster EXC 277 is a grant of the Deutsche

 * Forschungsgemeinschaft (DFG) in the context of the German

 * Excellence Initiative.

 *

 */

#pragma once

#include "AudioData.h"

#include <rst/audition/SoundChunk.pb.h>

#include <rst/audition/Utterance.pb.h>

using namespace std;

// converter from rsb utterance

class UtteranceRSB : public Utterance {

public:

        UtteranceRSB(rst::audition::Utterance rst_utterance) {

                // set text:

                set_text(rst_utterance.textual_representation());

                // convert soundchunk to audio data:

                extract_audio_data(rst_utterance.audio());

                // convert phonemes:

                extract_phonemes(rst_utterance);

        }

        ~UtteranceRSB(){};

        void extract_audio_data(rst::audition::SoundChunk sound_chunk) {

                // cout << sound_chunk.DebugString();

                // extract data:

                unsigned int audio_len = sound_chunk.data().length();

                char* audio_data_char = (char*)sound_chunk.data().c_str();

                // audio.samples = vector<char>(audio_data_char, audio_data_char+audio_len);

                audio_data->samples.resize(audio_len);

                audio_data->samples.assign(audio_data_char, audio_data_char + audio_len);

                printf("audio samplesize is %d bytes\n", (unsigned int)audio_data->samples.size());

                // extract format:

                audio_data->sample_signed = true;

                switch (sound_chunk.sample_type()) {

                        case (rst::audition::SoundChunk::SAMPLE_U8):

                                audio_data->sample_signed = false; // and fall through:

                        case (rst::audition::SoundChunk::SAMPLE_S8):

                                audio_data->sample_bit = 8;

                                break;

                        case (rst::audition::SoundChunk::SAMPLE_U16):

                                audio_data->sample_signed = false; // and fall through:

                        case (rst::audition::SoundChunk::SAMPLE_S16):

                                audio_data->sample_bit = 16;

                                break;

                        case (rst::audition::SoundChunk::SAMPLE_U24):

                                audio_data->sample_signed = false; // and fall through:

                        case (rst::audition::SoundChunk::SAMPLE_S24):

                                audio_data->sample_bit = 24;

                                break;

                        default:

                                printf("> invalid sample type %d in RST SoundChunk! ignoring request!\n", sound_chunk.sample_type());

                                throw runtime_error("UtteranceRSB::convert_audio_data() unsupported sample type in rst::audition::SoundChunk");

                }

                // bitrate

                audio_data->sample_rate = sound_chunk.rate();

                // endianness

                if (sound_chunk.endianness() == rst::audition::SoundChunk::ENDIAN_LITTLE) {

                        audio_data->sample_big_endian = false;

                }

                else if (sound_chunk.endianness() == rst::audition::SoundChunk::ENDIAN_BIG) {

                        audio_data->sample_big_endian = true;

                }

                else {

                        printf("> invalid SoundChunk byte_format");

                        throw runtime_error("UtteranceRSB::convert_audio_data() unsupported byte_format in rst::audition::SoundChunk");

                }

                // number of channels

                audio_data->sample_channels = sound_chunk.channels();

                printf("> new AudioData: %s\n", audio_data->to_string().c_str());

        }

        void extract_phonemes(rst::audition::Utterance rst_utterance) {

                // extract phoneme vector

                phonemes_vector.clear();

                for (int i = 0; i < rst_utterance.phonemes().element().size(); i++) {

                        rst::audition::Phoneme rst_phoneme = rst_utterance.phonemes().element().Get(i);

                        Utterance::symbol_duration_pair_t phoneme = make_pair(rst_phoneme.symbol(), rst_phoneme.duration());

                        phonemes_vector.push_back(phoneme);

                }

        }

};