0

我正在尝试使用 pocketsphinx 构建一个项目。我处于早期状态,我首先尝试了单个 main.c 中文档中的一个简单示例,该示例读取文件并检测单词:works。

现在我试图将它包含到我的 c++ SDL 项目中,通过 RtAudio 从麦克风读取音频数据,但它不起作用。

我得到:

INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 0 words
ERROR: "ngram_search.c", line 1141: Couldn't find <s> in first frame

麦克风初始化为:

RtAudio::StreamParameters parameters;
parameters.deviceId = adc.getDefaultInputDevice();
parameters.nChannels = 1;
parameters.firstChannel = 0;
unsigned int sampleRate = 16000;
unsigned int bufferFrames = 512;
adc.openStream(NULL, &parameters, RTAUDIO_SINT16, sampleRate, &bufferFrames, &rtCallback, info);
adc.startStream();

然后在我的 c++ 类中,我有以下受保护的变量:

cmd_ln_t *decoderConfig;
ps_decoder_t *currentDecoder;
bool spaceDown_; // true as long as user holds space
bool startNextTime_; // true if user just pressed space first time
bool endNextTime_; // true if user just released space
int16 *detectionBuffer;
int detectionBufferSize;
int detectionBufferPos;

在构造函数中我这样做(删除了错误检查以获得更好的可读性):

#define MODELDIR "/usr/local/share/pocketsphinx/model"
decoderConfig = cmd_ln_init(NULL, ps_args(), TRUE,
                    "-hmm", MODELDIR "/hmm/en_US/hub4wsj_sc_8k",
                    "-lm", MODELDIR "/lm/en/turtle.DMP",
                    "-dict", MODELDIR "/lm/en/turtle.dic",
                    NULL);
currentDecoder = ps_init(decoderConfig);

然后,当用户按下空格时,我将 startNextTime_ 和 spaceDown_ 设置为 true,如果他释放空间,我将 stopNextTime_ 设置为 true。

RtAudio 回调调用该类的方法,该方法执行以下操作(我将整个音频数据从用户按下空间复制到释放到缓冲区中,不确定这是否有必要,但我猜不会伤害):

if (spaceDown()) {
            if (startNextTime()) {
                    int rv = ps_start_utt(currentDecoder);
                    if (rv < 0) {
                            std::cout << "error on ps_start_utt" << std::endl;
                    }
                    setStartNextTime(false);
                    if (detectionBuffer != 0) {
                            free(detectionBuffer);
                    }
                    detectionBufferSize = 65536;
                    detectionBuffer = (int16*)malloc(detectionBufferSize*sizeof(int16));
                    detectionBufferPos = 0;
            }

            if (frames+detectionBufferPos > detectionBufferSize) {
                    detectionBufferSize *= 2;
                    detectionBuffer = (int16*)realloc(detectionBuffer, detectionBufferSize*sizeof(int16));
            }

            memcpy(detectionBuffer+detectionBufferPos, buf, frames*sizeof(int16));

            ps_process_raw(currentDecoder, detectionBuffer+detectionBufferPos, (size_t)frames, 0, 1);
            detectionBufferPos += frames;
            if (endNextTime()) {
                    int rv = ps_end_utt(currentDecoder);
                    int32 score = 0;
                    char const *hyp = ps_get_hyp(currentDecoder, &score);
                    if (hyp != NULL) {
                            std::cout << "got " << hyp << " with score " << score << " and prob " << ps_get_prob(currentDecoder) << std::endl;;
                    } else {
                            std::cout << "no hyp " << std::endl;
                    }
                    setSpaceDown(false);
                    setEndNextTime(false);
            }
}

在上面的 pocketsphinx 错误消息之后,我得到了“no hyp”输出。我一遍又一遍地将它与我的小型测试 c 文件进行比较,唯一的区别是 a)我正在从麦克风而不是文件中读取数据,并且 b)它在线程中运行。

有任何想法吗?

编辑:

这是pocketsphinx日志:

    INFO: cmd_ln.c(697): Parsing command line:
\
    -hmm /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k \
    -lm /usr/local/share/pocketsphinx/model/lm/en/turtle.DMP \
    -dict /usr/local/share/pocketsphinx/model/lm/en/turtle.dic 

Current configuration:
[NAME]      [DEFLT]     [VALUE]
-agc        none        none
-agcthresh  2.0     2.000000e+00
-allphone           
-allphone_ci    no      no
-alpha      0.97        9.700000e-01
-ascale     20.0        2.000000e+01
-aw     1       1
-backtrace  no      no
-beam       1e-48       1.000000e-48
-bestpath   yes     yes
-bestpathlw 9.5     9.500000e+00
-ceplen     13      13
-cmn        current     current
-cmninit    8.0     8.0
-compallsen no      no
-debug              0
-dict               /usr/local/share/pocketsphinx/model/lm/en/turtle.dic
-dictcase   no      no
-dither     no      no
-doublebw   no      no
-ds     1       1
-fdict              
-feat       1s_c_d_dd   1s_c_d_dd
-featparams         
-fillprob   1e-8        1.000000e-08
-frate      100     100
-fsg                
-fsgusealtpron  yes     yes
-fsgusefiller   yes     yes
-fwdflat    yes     yes
-fwdflatbeam    1e-64       1.000000e-64
-fwdflatefwid   4       4
-fwdflatlw  8.5     8.500000e+00
-fwdflatsfwin   25      25
-fwdflatwbeam   7e-29       7.000000e-29
-fwdtree    yes     yes
-hmm                /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k
-input_endian   little      little
-jsgf               
-keyphrase          
-kws                
-kws_plp    1e-1        1.000000e-01
-kws_threshold  1       1.000000e+00
-latsize    5000        5000
-lda                
-ldadim     0       0
-lifter     0       0
-lm             /usr/local/share/pocketsphinx/model/lm/en/turtle.DMP
-lmctl              
-lmname             
-logbase    1.0001      1.000100e+00
-logfn              
-logspec    no      no
-lowerf     133.33334   1.333333e+02
-lpbeam     1e-40       1.000000e-40
-lponlybeam 7e-29       7.000000e-29
-lw     6.5     6.500000e+00
-maxhmmpf   30000       30000
-maxwpf     -1      -1
-mdef               
-mean               
-mfclogdir          
-min_endfr  0       0
-mixw               
-mixwfloor  0.0000001   1.000000e-07
-mllr               
-mmap       yes     yes
-ncep       13      13
-nfft       512     512
-nfilt      40      40
-nwpen      1.0     1.000000e+00
-pbeam      1e-48       1.000000e-48
-pip        1.0     1.000000e+00
-pl_beam    1e-10       1.000000e-10
-pl_pbeam   1e-10       1.000000e-10
-pl_pip     1.0     1.000000e+00
-pl_weight  3.0     3.000000e+00
-pl_window  5       5
-rawlogdir          
-remove_dc  no      no
-remove_noise   yes     yes
-remove_silence yes     yes
-round_filters  yes     yes
-samprate   16000       1.600000e+04
-seed       -1      -1
-sendump            
-senlogdir          
-senmgau            
-silprob    0.005       5.000000e-03
-smoothspec no      no
-svspec             
-tmat               
-tmatfloor  0.0001      1.000000e-04
-topn       4       4
-topn_beam  0       0
-toprule            
-transform  legacy      legacy
-unit_area  yes     yes
-upperf     6855.4976   6.855498e+03
-uw     1.0     1.000000e+00
-vad_postspeech 50      50
-vad_prespeech  10      10
-vad_threshold  2.0     2.000000e+00
-var                
-varfloor   0.0001      1.000000e-04
-varnorm    no      no
-verbose    no      no
-warp_params            
-warp_type  inverse_linear  inverse_linear
-wbeam      7e-29       7.000000e-29
-wip        0.65        6.500000e-01
-wlen       0.025625    2.562500e-02

INFO: cmd_ln.c(697): Parsing command line:
\
    -nfilt 20 \
    -lowerf 1 \
    -upperf 4000 \
    -wlen 0.025 \
    -transform dct \
    -round_filters no \
    -remove_dc yes \
    -remove_noise no \
    -svspec 0-12/13-25/26-38 \
    -feat 1s_c_d_dd \
    -agc none \
    -cmn current \
    -cmninit 45,-3,1 \
    -varnorm no 

Current configuration:
[NAME]      [DEFLT]     [VALUE]
-agc        none        none
-agcthresh  2.0     2.000000e+00
-alpha      0.97        9.700000e-01
-ceplen     13      13
-cmn        current     current
-cmninit    8.0     45,-3,1
-dither     no      no
-doublebw   no      no
-feat       1s_c_d_dd   1s_c_d_dd
-frate      100     100
-input_endian   little      little
-lda                
-ldadim     0       0
-lifter     0       0
-logspec    no      no
-lowerf     133.33334   1.000000e+00
-ncep       13      13
-nfft       512     512
-nfilt      40      20
-remove_dc  no      yes
-remove_noise   yes     no
-remove_silence yes     yes
-round_filters  yes     no
-samprate   16000       1.600000e+04
-seed       -1      -1
-smoothspec no      no
-svspec             0-12/13-25/26-38
-transform  legacy      dct
-unit_area  yes     yes
-upperf     6855.4976   4.000000e+03
-vad_postspeech 50      50
-vad_prespeech  10      10
-vad_threshold  2.0     2.000000e+00
-varnorm    no      no
-verbose    no      no
-warp_params            
-warp_type  inverse_linear  inverse_linear
-wlen       0.025625    2.500000e-02

INFO: acmod.c(252): Parsed model-specific feature parameters from /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/feat.params
INFO: feat.c(715): Initializing feature stream to type: '1s_c_d_dd', ceplen=13, CMN='current', VARNORM='no', AGC='none'
INFO: cmn.c(143): mean[0]= 12.00, mean[1..12]= 0.0
INFO: acmod.c(171): Using subvector specification 0-12/13-25/26-38
INFO: mdef.c(518): Reading model definition: /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/mdef
INFO: mdef.c(531): Found byte-order mark BMDF, assuming this is a binary mdef file
INFO: bin_mdef.c(336): Reading binary model definition: /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/mdef
INFO: bin_mdef.c(516): 50 CI-phone, 143047 CD-phone, 3 emitstate/phone, 150 CI-sen, 5150 Sen, 27135 Sen-Seq
INFO: tmat.c(206): Reading HMM transition probability matrices: /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/transition_matrices
INFO: acmod.c(124): Attempting to use PTM computation module
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/means
INFO: ms_gauden.c(292): 1 codebook, 3 feature, size: 
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/variances
INFO: ms_gauden.c(292): 1 codebook, 3 feature, size: 
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(354): 0 variance values floored
INFO: ptm_mgau.c(805): Number of codebooks doesn't match number of ciphones, doesn't look like PTM: 1 != 50
INFO: acmod.c(126): Attempting to use semi-continuous computation module
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/means
INFO: ms_gauden.c(292): 1 codebook, 3 feature, size: 
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(198): Reading mixture gaussian parameter: /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/variances
INFO: ms_gauden.c(292): 1 codebook, 3 feature, size: 
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(294):  256x13
INFO: ms_gauden.c(354): 0 variance values floored
INFO: s2_semi_mgau.c(904): Loading senones from dump file /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/sendump
INFO: s2_semi_mgau.c(928): BEGIN FILE FORMAT DESCRIPTION
INFO: s2_semi_mgau.c(1023): Using memory-mapped I/O for senones
INFO: s2_semi_mgau.c(1294): Maximum top-N: 4 Top-N beams: 0 0 0
INFO: phone_loop_search.c(115): State beam -225 Phone exit beam -225 Insertion penalty 0
INFO: dict.c(320): Allocating 4217 * 32 bytes (131 KiB) for word entries
INFO: dict.c(333): Reading main dictionary: /usr/local/share/pocketsphinx/model/lm/en/turtle.dic
INFO: dict.c(213): Allocated 0 KiB for strings, 0 KiB for phones
INFO: dict.c(336): 110 words read
INFO: dict.c(342): Reading filler dictionary: /usr/local/share/pocketsphinx/model/hmm/en_US/hub4wsj_sc_8k/noisedict
INFO: dict.c(213): Allocated 0 KiB for strings, 0 KiB for phones
INFO: dict.c(345): 11 words read
INFO: dict2pid.c(396): Building PID tables for dictionary
INFO: dict2pid.c(406): Allocating 50^3 * 2 bytes (244 KiB) for word-initial triphones
INFO: dict2pid.c(132): Allocated 60400 bytes (58 KiB) for word-final triphones
INFO: dict2pid.c(196): Allocated 60400 bytes (58 KiB) for single-phone word triphones
INFO: ngram_model_arpa.c(77): No \data\ mark in LM file
INFO: ngram_model_dmp.c(142): Will use memory-mapped I/O for LM file
INFO: ngram_model_dmp.c(196): ngrams 1=91, 2=212, 3=177
INFO: ngram_model_dmp.c(242):       91 = LM.unigrams(+trailer) read
INFO: ngram_model_dmp.c(288):      212 = LM.bigrams(+trailer) read
INFO: ngram_model_dmp.c(314):      177 = LM.trigrams read
INFO: ngram_model_dmp.c(339):       20 = LM.prob2 entries read
INFO: ngram_model_dmp.c(359):       12 = LM.bo_wt2 entries read
INFO: ngram_model_dmp.c(379):       12 = LM.prob3 entries read
INFO: ngram_model_dmp.c(407):        1 = LM.tseg_base entries read
INFO: ngram_model_dmp.c(463):       91 = ascii word strings read
INFO: ngram_search_fwdtree.c(99): 67 unique initial diphones
INFO: ngram_search_fwdtree.c(148): 0 root, 0 non-root channels, 15 single-phone words
INFO: ngram_search_fwdtree.c(186): Creating search tree
INFO: ngram_search_fwdtree.c(192): before: 0 root, 0 non-root channels, 15 single-phone words
INFO: ngram_search_fwdtree.c(326): after: max nonroot chan increased to 328
INFO: ngram_search_fwdtree.c(339): after: 67 root, 200 non-root channels, 14 single-phone words
INFO: ngram_search_fwdflat.c(157): fwdflat: min_ef_width = 4, max_sf_win = 25
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 0 words
ERROR: "ngram_search.c", line 1141: Couldn't find <s> in first frame
no hyp
4

1 回答 1

0

我有类似的经历(解析文件工作正常;切换到麦克风会出现错误"Couldn't find <s> in first frame")。我检查了麦克风数据流,发现它有一个 WAV 标头。我将麦克风流切换到 RAW 而不是默认的 WAV,它开始工作。我对 RtAudio 不熟悉,但我认为:1)它不是原始 PCM 数据,或者 2)它以标头开头,狮身人面像不喜欢。

注意:我正在使用arecordRaspbian Jessie 在 Raspberry Pi 上获取麦克风流。

于 2016-07-21T21:29:40.040 回答