Revision 42d7a3f5

View differences:

server/action/tts.action
1
#utterance
2
string text
3

  
4
---
5
#response
6
utterance utterance
7

  
8
---
9
uint32 result
10

  
server/include/ROS/UtteranceROS.h
1
/*
2
* This file is part of hlrc_server
3
*
4
* Copyright(c) sschulz <AT> techfak.uni-bielefeld.de
5
* http://opensource.cit-ec.de/projects/hlrc_server
6
*
7
* This file may be licensed under the terms of the
8
* GNU General Public License Version 3 (the ``GPL''),
9
* or (at your option) any later version.
10
*
11
* Software distributed under the License is distributed
12
* on an ``AS IS'' basis, WITHOUT WARRANTY OF ANY KIND, either
13
* express or implied. See the GPL for the specific language
14
* governing rights and limitations.
15
*
16
* You should have received a copy of the GPL along with this
17
* program. If not, go to http://www.gnu.org/licenses/gpl.html
18
* or write to the Free Software Foundation, Inc.,
19
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
20
*
21
* The development of this software was supported by the
22
* Excellence Cluster EXC 277 Cognitive Interaction Technology.
23
* The Excellence Cluster EXC 277 is a grant of the Deutsche
24
* Forschungsgemeinschaft (DFG) in the context of the German
25
* Excellence Initiative.
26
*
27
*/
28

  
29
#pragma once
30
#include "AudioData.h"
31
using namespace std;
32
using namespace hlrc_server;
33
#include "hlrc_server/phoneme.h"
34
#include "hlrc_server/soundchunk.h"
35
#include "hlrc_server/utterance.h"
36
#include "hlrc_server/ttsAction.h"
37

  
38
//converter from ros utterance
39
class UtteranceROS : public Utterance {
40
    public:
41
        UtteranceROS(ttsResultConstPtr ros_utterance){
42
            //set text:
43
            set_text(ros_utterance->utterance.text);
44

  
45
            //convert soundchunk to audio data:
46
            extract_audio_data(ros_utterance->utterance.audio);
47

  
48
            //convert phonemes:
49
            extract_phonemes(ros_utterance->utterance.phonemes);
50
        }
51

  
52
        ~UtteranceROS(){};
53

  
54
        void extract_audio_data(soundchunk sound_chunk){
55
            //extract data:
56
            unsigned int audio_len = sound_chunk.data.size();
57
            char *audio_data_char = (char *)sound_chunk.data.data();
58

  
59
            audio_data->samples.resize(audio_len);
60
            audio_data->samples.assign(audio_data_char, audio_data_char+audio_len);
61
            printf("audio samplesize is %d bytes\n",(unsigned int)audio_data->samples.size());
62

  
63
            /*
64
             * : data()
65
            , samplecount(0)
66
            , channels(0)
67
            , rate(0)
68
            , sample_type(0)
69
            , endianess(0)  {
70

  
71
            }*/
72

  
73
            //extract format:
74
            audio_data->sample_signed = true;
75
            switch (sound_chunk.sample_type){
76
                case(soundchunk::SAMPLE_U8):  audio_data->sample_signed = false; //and fall through:
77
                case(soundchunk::SAMPLE_S8):  audio_data->sample_bit =  8; break;
78

  
79
                case(soundchunk::SAMPLE_U16): audio_data->sample_signed = false; //and fall through:
80
                case(soundchunk::SAMPLE_S16): audio_data->sample_bit = 16; break;
81

  
82
                case(soundchunk::SAMPLE_U24): audio_data->sample_signed = false; //and fall through:
83
                case(soundchunk::SAMPLE_S24): audio_data->sample_bit = 24; break;
84

  
85
                default:
86
                    printf("> invalid sample type %d in ROS SoundChunk! ignoring request!\n", sound_chunk.sample_type);
87
                    throw runtime_error("UtteranceROS::convert_audio_data() unsupported sample type in ros SoundChunk");
88
            }
89

  
90
            //bitrate
91
            audio_data->sample_rate = sound_chunk.rate;
92

  
93
            //endianness
94
            if (sound_chunk.endianess == soundchunk::ENDIAN_LITTLE){
95
                audio_data->sample_big_endian = false;
96
            }else if (sound_chunk.endianess == soundchunk::ENDIAN_BIG){
97
                audio_data->sample_big_endian = true;
98
            }else{
99
                printf("> invalid SoundChunk byte_format");
100
                throw runtime_error("UtteranceROS::convert_audio_data() unsupported byte_format in ros SoundChunk");
101
            }
102

  
103
        //number of channels
104
        audio_data->sample_channels = sound_chunk.channels;
105

  
106
        printf("> new AudioData: %s\n",audio_data->to_string().c_str());
107
    }
108

  
109
    void extract_phonemes(utterance::_phonemes_type ros_phones){
110
        //extract phoneme vector
111
        phonemes_vector.clear();
112
        for(auto it = ros_phones.begin(); it<ros_phones.end(); it++){
113
            Utterance::symbol_duration_pair_t phoneme = make_pair(it->symbol, it->duration);
114
            phonemes_vector.push_back(phoneme);
115
        }
116
    }
117

  
118
};
119

  
server/msg/utterance.msg
1
#list of symbol/duration pairs
2
phoneme[] phonemes
3

  
4
#sound data (raw data, similar to a wave file)
5
soundchunk audio
6

  
7
#textual description
8
string text
9

  

Also available in: Unified diff