In [1]:

    

# All the audio processing tools are wrapped in the Python Module called AudioPipe
import AudioPipe.speaker.recognition as SR # Speaker Recognition Module
import AudioPipe.fingerprint.panako as FP # Accoustic Fingerprinting Module
from AudioPipe.speaker.silence import remove_silence # tool for remove the silence in the audio, not needed
import numpy as np
from AudioPipe.features import mfcc # Feature Extraction Module, part of the shared preprocessing
import scipy.io.wavfile as wav 
from AudioPipe.speaker.rec import dia2spk, spk # Speaker Recognition using diarization results
from AudioPipe.utils.utils import video2audio # Format converting module, part of the shared preprocessing
import commands, os
from AudioPipe.diarization.diarization import Diarization # Speaker Diarization Moudule
import AudioPipe.data.manage as DM # Data Management Module


    

In [2]:

    

# Select the video file to be processed
Video_node = DM.Node("Data/Video/",".mp4")
name = "2015-08-07_0050_US_FOX-News_US_Presidential_Politics"


    

    




(0, '')

In [3]:

    

# Select the file for the meta infomation
Meta_node = DM.Node("Data/RedHen/",".seg")
meta = Meta_node.Pick(name)


    

    




(0, '')

In [4]:

    

# Store the fingerprint of the video
FP_node= DM.Node("Data/Fingerprint/")
output, err, exitcode = Video_node.Flow(FP.Store, name, FP_node, [])


    

    




(0, '')

In [5]:

    

# Convert the video to audio
Audio_node = DM.Node("Data/Audio/", ".wav")
audio = Video_node.Flow(video2audio, name, Audio_node, [Audio_node.ext])


    

    




(0, '')

In [6]:

    

Dia_node = DM.Node("Data/Diarization/", ".rttm")
args = dict(init_cluster=20, dest_mfcc='Data/MFCC', dest_cfg="Data/Model/DiaCfg")
dia =  Audio_node.Flow(Diarization, name, Dia_node, args)


    

    




(0, '')
(0, '')
(0, '')
(0, '')
(0, '')
(0, '')
Now start extracting mfcc features
(0, "ffmpeg version 2.8.2 Copyright (c) 2000-2015 the FFmpeg developers\n  built with gcc 4.9.3 (GCC)\n  configuration: --prefix=/usr/local/ffmpeg/2.8.2 --enable-shared\n  libavutil      54. 31.100 / 54. 31.100\n  libavcodec     56. 60.100 / 56. 60.100\n  libavformat    56. 40.101 / 56. 40.101\n  libavdevice    56.  4.100 / 56.  4.100\n  libavfilter     5. 40.101 /  5. 40.101\n  libswscale      3.  1.101 /  3.  1.101\n  libswresample   1.  2.101 /  1.  2.101\nGuessed Channel Layout for  Input Stream #0.0 : stereo\nInput #0, wav, from 'Data/Audio/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.wav':\n  Metadata:\n    encoder         : Lavf56.40.101\n  Duration: 02:09:53.41, bitrate: 1411 kb/s\n    Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 44100 Hz, 2 channels, s16, 1411 kb/s\nOutput #0, wav, to '/home/hxx124/Pipeline/Audio/Pipeline/AudioPipe/diarization/temp_audio/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.wav':\n  Metadata:\n    ISFT            : Lavf56.40.101\n    Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 16000 Hz, stereo, s16, 512 kb/s\n    Metadata:\n      encoder         : Lavc56.60.100 pcm_s16le\nStream mapping:\n  Stream #0:0 -> #0:0 (pcm_s16le (native) -> pcm_s16le (native))\nPress [q] to stop, [?] for help\nsize=   14354kB time=00:03:49.66 bitrate= 512.0kbits/s    \rsize=   30254kB time=00:08:04.06 bitrate= 512.0kbits/s    \rsize=   46187kB time=00:12:18.99 bitrate= 512.0kbits/s    \rsize=   62105kB time=00:16:33.67 bitrate= 512.0kbits/s    \rsize=   74326kB time=00:19:49.20 bitrate= 512.0kbits/s    \rsize=   90270kB time=00:24:04.32 bitrate= 512.0kbits/s    \rsize=  106214kB time=00:28:19.42 bitrate= 512.0kbits/s    \rsize=  122199kB time=00:32:35.18 bitrate= 512.0kbits/s    \rsize=  134161kB time=00:35:46.56 bitrate= 512.0kbits/s    \rsize=  150089kB time=00:40:01.42 bitrate= 512.0kbits/s    \rsize=  166008kB time=00:44:16.12 bitrate= 512.0kbits/s    \rsize=  181912kB time=00:48:30.59 bitrate= 512.0kbits/s    \rsize=  196599kB time=00:52:25.58 bitrate= 512.0kbits/s    \rsize=  212455kB time=00:56:39.28 bitrate= 512.0kbits/s    \rsize=  228341kB time=01:00:53.44 bitrate= 512.0kbits/s    \rsize=  244223kB time=01:05:07.56 bitrate= 512.0kbits/s    \rsize=  260149kB time=01:09:22.38 bitrate= 512.0kbits/s    \rsize=  271652kB time=01:12:26.42 bitrate= 512.0kbits/s    \rsize=  287584kB time=01:16:41.33 bitrate= 512.0kbits/s    \rsize=  303520kB time=01:20:56.31 bitrate= 512.0kbits/s    \rsize=  319472kB time=01:25:11.54 bitrate= 512.0kbits/s    \rsize=  331968kB time=01:28:31.49 bitrate= 512.0kbits/s    \rsize=  347867kB time=01:32:45.86 bitrate= 512.0kbits/s    \rsize=  363761kB time=01:37:00.17 bitrate= 512.0kbits/s    \rsize=  379690kB time=01:41:15.03 bitrate= 512.0kbits/s    \rsize=  393195kB time=01:44:51.12 bitrate= 512.0kbits/s    \rsize=  409082kB time=01:49:05.30 bitrate= 512.0kbits/s    \rsize=  425012kB time=01:53:20.19 bitrate= 512.0kbits/s    \rsize=  440989kB time=01:57:35.82 bitrate= 512.0kbits/s    \rsize=  456941kB time=02:01:51.05 bitrate= 512.0kbits/s    \rsize=  466085kB time=02:04:17.36 bitrate= 512.0kbits/s    \rsize=  482049kB time=02:08:32.78 bitrate= 512.0kbits/s    \rsize=  487088kB time=02:09:53.40 bitrate= 512.0kbits/s    \nvideo:0kB audio:487088kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.000016%")
(0, '')
Now start preparing configure file
Now start clustering
(0, "/usr/bin/ld: skipping incompatible /usr/local/cuda-7.0/lib/libcudart.so when searching for -lcudart\n/usr/bin/ld: skipping incompatible /usr/local/cuda-7.0/lib/libcudart.so when searching for -lcudart\ng++ -pthread -fno-strict-aliasing -g -O2 -g -fwrapv -O3 -Wall -fPIC -pthread -shared -Xlinker -export-dynamic -DNDEBUG -I/usr/local/python/2.7.8/include/python2.7 -I/usr/local/cuda-7.0/include /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/46c837d5cb878372cdb2886923bd37a6/module.o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/7ceed7a1666bbc7019867d851e0660de/gpu.o -L/usr/local/python/2.7.8/lib -L/usr/local/cuda-7.0/lib -L/usr/local/cuda-7.0/lib64 -lcuda -lcudart -lboost_python -lpython2.7 -lpthread -ldl -lutil -o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/codepy.temp.46c837d5cb878372cdb2886923bd37a6.7ceed7a1666bbc7019867d851e0660de.module.so\nUsing the config file specified: ' Data/Model/DiaCfg/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.cfg '\nSpeech file spnsp_file not specified, continuing without it...\n...Reading in HTK feature file...\nINFO: total number of frames read:  779339\n...Reading in speech/nonspeech file...\n ====================== CLUSTERING ON SUBSET ====================== \nGMM SPECIALIZER: USING CUDA\ng++ -pthread -fno-strict-aliasing -g -O2 -g -fwrapv -O3 -Wall -fPIC -pthread -shared -Xlinker -export-dynamic -DNDEBUG -I/usr/local/python/2.7.8/include/python2.7 -I/home/hxx124/myPython/virtualenv-1.9/ENV/lib/python2.7/site-packages/numpy/core/include -I/usr/local/cuda-7.0/include /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/38d44195c321f1de7e40420e316a4733/module.o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.o -L/usr/local/python/2.7.8/lib -L/usr/local/cuda-7.0/lib -L/usr/local/cuda-7.0/lib64 -lcuda -lcudart -lboost_python -lpython2.7 -lpthread -ldl -lutil -o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/codepy.temp.38d44195c321f1de7e40420e316a4733.11dee452985a0f3b9b6a40453d090df4.module.so\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 15]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 15, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 15, 10, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 15, 10, 15]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 10, 10, 15, 10, 20]\n size of each cluster: [5, 5, 5, 5, 5, 5, 10, 15, 10, 20, 15]\n size of each cluster: [5, 5, 5, 5, 5, 5, 10, 15, 10, 35]\n size of each cluster: [5, 5, 5, 5, 5, 15, 10, 35, 15]\n size of each cluster: [5, 5, 5, 5, 10, 35, 15, 20]\n size of each cluster: [5, 5, 5, 35, 15, 20, 15]\n size of each cluster: [5, 5, 35, 15, 20, 20]\n size of each cluster: [5, 5, 35, 20, 35]\n size of each cluster: [5, 5, 35, 55]\n size of each cluster: [5, 55, 40]\n size of each cluster: [40, 60]\n size of each cluster: [100]\n=== Total clustering time:  935.101713181\n=== Final size of each cluster: [100]\n...Writing out RTTM file...\nDONE writing RTTM file")
(0, '')

In [7]:

    

# Below you can find an example on training a fresh model
# However, please make sure that the sample rate of your training data is the same with your testing data
model_gender_dir = "Data/Model/Gender/"
model_gender_nm = 'gender.model'
model_gender = model_gender_dir+model_gender_nm
if not os.path.isfile(model_gender):
    Gender = SR.GMMRec() # Create a new recognizer
    female_fn = model_gender_dir+'female.wav' # choose the training file for female
    male_fn = model_gender_dir+'male.wav' # choose the training file for male
    Gender.enroll_file('Female', female_fn) # enroll the female audio 
    Gender.enroll_file('Male', male_fn) # enroll the male audio 
    Gender.train() # train the GMMs after you enroll all the training data 
    Gender.dump(model_gender) # save the trained model into a file named "gender.model" for future use


    

In [ ]:

    

# Gender Identification based on Speaker Diarization
Gen_node = DM.Node("Data/Gender/",".gen")
gen = Audio_node.Flow(dia2spk, name, Gen_node, [model_gender, dia, meta, Gen_node.ext])


    

In [ ]:

    

Genr_node = DM.Node("Data/Gender/",".genr")
gen = Audio_node.Flow(spk, name, Genr_node, [model_gender, dia, meta, Genr_node.ext])


    

In [ ]:

    

# you can train a fresh model as follows:

model_speaker_dir = "Data/Model/Speaker/" # The training data should already be here!! 
model_speaker_nm = 'speaker.model'
model_speaker = model_speaker_dir+model_speaker_nm
if not os.path.isfile(model_speaker):
    Speaker = SR.GMMRec() # Create a new recognizer
    other_fn = model_speaker_dir+'Imposter.wav' # choose the training file for female
    trump_fn = model_speaker_dir+'Trump.wav' # choose the training file for male
    Speaker.enroll_file('Other', other_fn) # enroll the female audio 
    Speaker.enroll_file('Trump', trump_fn) # enroll the male audio 
    # You can add more speakers here following the syntax above
    Speaker.train() # train the GMMs after you enroll all the training data 
    Speaker.dump(model_speaker) # save the trained model into a file named "gender.model" for future use


    

In [ ]:

    




    

In [2]:

    

# Select the video file to be processed
video = "Data/Video/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.mp4"
name = os.path.basename(video).split('.')[0]


    

In [3]:

    

# Select the file for the meta infomation
meta = "Data/RedHen/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.seg"


    

In [4]:

    

# Store the fingerprint of the video
fingerprint_dir = "Data/Fingerprint/"
commands.getstatusoutput("mkdir -p "+fingerprint_dir)
output, err, exitcode = FP.Store(video, fingerprint_dir)


    

In [5]:

    

# Convert the video to audio
audio_dir = "Data/Audio/"
commands.getstatusoutput("mkdir -p "+audio_dir)
audio_ext = ".wav"
video2audio(video, audio_dir, audio_ext)
audio = os.path.join(audio_dir, name+audio_ext)


    

In [6]:

    

# Speaker Diarization
dia_dir = "Data/Diarization/"
commands.getstatusoutput("mkdir -p "+dia_dir)
dia_ext = ".rttm"
Diarization(audio, dest_cfg="Data/Model/DiaCfg", dest_rttm=dia_dir, dest_mfcc="Data/MFCC/")
dia = os.path.join(dia_dir,name+dia_ext)


    

    




(0, '')
(0, '')
(0, '')
(0, '')
(0, '')
(0, "ffmpeg version 2.8.2 Copyright (c) 2000-2015 the FFmpeg developers\n  built with gcc 4.9.3 (GCC)\n  configuration: --prefix=/usr/local/ffmpeg/2.8.2 --enable-shared\n  libavutil      54. 31.100 / 54. 31.100\n  libavcodec     56. 60.100 / 56. 60.100\n  libavformat    56. 40.101 / 56. 40.101\n  libavdevice    56.  4.100 / 56.  4.100\n  libavfilter     5. 40.101 /  5. 40.101\n  libswscale      3.  1.101 /  3.  1.101\n  libswresample   1.  2.101 /  1.  2.101\nGuessed Channel Layout for  Input Stream #0.0 : stereo\nInput #0, wav, from 'Data/Audio/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.wav':\n  Metadata:\n    encoder         : Lavf56.40.101\n  Duration: 02:09:53.41, bitrate: 1411 kb/s\n    Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 44100 Hz, 2 channels, s16, 1411 kb/s\nOutput #0, wav, to '/home/hxx124/Pipeline/Audio/Pipeline/AudioPipe/diarization/temp_audio/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.wav':\n  Metadata:\n    ISFT            : Lavf56.40.101\n    Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 16000 Hz, stereo, s16, 512 kb/s\n    Metadata:\n      encoder         : Lavc56.60.100 pcm_s16le\nStream mapping:\n  Stream #0:0 -> #0:0 (pcm_s16le (native) -> pcm_s16le (native))\nPress [q] to stop, [?] for help\nsize=    7160kB time=00:01:54.56 bitrate= 512.0kbits/s    \rsize=   15074kB time=00:04:01.18 bitrate= 512.0kbits/s    \rsize=   22875kB time=00:06:05.99 bitrate= 512.0kbits/s    \rsize=   30791kB time=00:08:12.65 bitrate= 512.0kbits/s    \rsize=   44452kB time=00:11:51.22 bitrate= 512.0kbits/s    \rsize=   60381kB time=00:16:06.08 bitrate= 512.0kbits/s    \rsize=   68950kB time=00:18:23.20 bitrate= 512.0kbits/s    \rsize=   76890kB time=00:20:30.23 bitrate= 512.0kbits/s    \rsize=   84801kB time=00:22:36.81 bitrate= 512.0kbits/s    \rsize=   92649kB time=00:24:42.38 bitrate= 512.0kbits/s    \rsize=  100602kB time=00:26:49.62 bitrate= 512.0kbits/s    \rsize=  108463kB time=00:28:55.41 bitrate= 512.0kbits/s    \rsize=  116386kB time=00:31:02.16 bitrate= 512.0kbits/s    \rsize=  124250kB time=00:33:07.99 bitrate= 512.0kbits/s    \rsize=  131068kB time=00:34:57.08 bitrate= 512.0kbits/s    \rsize=  139128kB time=00:37:06.04 bitrate= 512.0kbits/s    \rsize=  147161kB time=00:39:14.57 bitrate= 512.0kbits/s    \rsize=  155014kB time=00:41:20.21 bitrate= 512.0kbits/s    \rsize=  162926kB time=00:43:26.81 bitrate= 512.0kbits/s    \rsize=  170746kB time=00:45:31.94 bitrate= 512.0kbits/s    \rsize=  178733kB time=00:47:39.72 bitrate= 512.0kbits/s    \rsize=  186587kB time=00:49:45.38 bitrate= 512.0kbits/s    \rsize=  194564kB time=00:51:53.02 bitrate= 512.0kbits/s    \rsize=  200774kB time=00:53:32.38 bitrate= 512.0kbits/s    \rsize=  208628kB time=00:55:38.05 bitrate= 512.0kbits/s    \rsize=  215799kB time=00:57:32.78 bitrate= 512.0kbits/s    \rsize=  227194kB time=01:00:35.10 bitrate= 512.0kbits/s    \rsize=  243111kB time=01:04:49.78 bitrate= 512.0kbits/s    \rsize=  259072kB time=01:09:05.15 bitrate= 512.0kbits/s    \rsize=  267019kB time=01:11:12.30 bitrate= 512.0kbits/s    \rsize=  275071kB time=01:13:21.13 bitrate= 512.0kbits/s    \rsize=  282932kB time=01:15:26.91 bitrate= 512.0kbits/s    \rsize=  290830kB time=01:17:33.27 bitrate= 512.0kbits/s    \rsize=  298680kB time=01:19:38.87 bitrate= 512.0kbits/s    \rsize=  306737kB time=01:21:47.79 bitrate= 512.0kbits/s    \rsize=  314555kB time=01:23:52.87 bitrate= 512.0kbits/s    \rsize=  322492kB time=01:25:59.86 bitrate= 512.0kbits/s    \rsize=  329147kB time=01:27:46.35 bitrate= 512.0kbits/s    \rsize=  337160kB time=01:29:54.55 bitrate= 512.0kbits/s    \rsize=  345040kB time=01:32:00.63 bitrate= 512.0kbits/s    \rsize=  353200kB time=01:34:11.20 bitrate= 512.0kbits/s    \rsize=  361075kB time=01:36:17.19 bitrate= 512.0kbits/s    \rsize=  369017kB time=01:38:24.27 bitrate= 512.0kbits/s    \rsize=  376889kB time=01:40:30.22 bitrate= 512.0kbits/s    \rsize=  384734kB time=01:42:35.74 bitrate= 512.0kbits/s    \rsize=  392512kB time=01:44:40.18 bitrate= 512.0kbits/s    \rsize=  398880kB time=01:46:22.07 bitrate= 512.0kbits/s    \rsize=  406828kB time=01:48:29.24 bitrate= 512.0kbits/s    \rsize=  414666kB time=01:50:34.65 bitrate= 512.0kbits/s    \rsize=  422570kB time=01:52:41.11 bitrate= 512.0kbits/s    \rsize=  430388kB time=01:54:46.20 bitrate= 512.0kbits/s    \rsize=  438394kB time=01:56:54.30 bitrate= 512.0kbits/s    \rsize=  446186kB time=01:58:58.97 bitrate= 512.0kbits/s    \rsize=  454024kB time=02:01:04.38 bitrate= 512.0kbits/s    \rsize=  460366kB time=02:02:45.85 bitrate= 512.0kbits/s    \rsize=  468213kB time=02:04:51.40 bitrate= 512.0kbits/s    \rsize=  476050kB time=02:06:56.79 bitrate= 512.0kbits/s    \rsize=  487088kB time=02:09:53.40 bitrate= 512.0kbits/s    \nvideo:0kB audio:487088kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.000016%")
(0, '')
Now start preparing configure file
Now start clustering
(0, '/usr/bin/ld: skipping incompatible /usr/local/cuda-7.0/lib/libcudart.so when searching for -lcudart\nIn file included from /home/hxx124/myPython/virtualenv-1.9/ENV/lib/python2.7/site-packages/numpy/core/include/numpy/ndarraytypes.h:1777,\n                 from /home/hxx124/myPython/virtualenv-1.9/ENV/lib/python2.7/site-packages/numpy/core/include/numpy/ndarrayobject.h:18,\n                 from /home/hxx124/myPython/virtualenv-1.9/ENV/lib/python2.7/site-packages/numpy/core/include/numpy/arrayobject.h:4,\n                 from /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/38d44195c321f1de7e40420e316a4733/module.cpp:9:\n/home/hxx124/myPython/virtualenv-1.9/ENV/lib/python2.7/site-packages/numpy/core/include/numpy/npy_1_7_deprecated_api.h:15:2: warning: #warning "Using deprecated NumPy API, disable it by " "#defining NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION"\ng++ -pthread -fno-strict-aliasing -g -O2 -g -fwrapv -O3 -Wall -fPIC -pthread -shared -Xlinker -export-dynamic -DNDEBUG -I/usr/local/python/2.7.8/include/python2.7 -I/usr/local/cuda-7.0/include /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/46c837d5cb878372cdb2886923bd37a6/module.o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/7ceed7a1666bbc7019867d851e0660de/gpu.o -L/usr/local/python/2.7.8/lib -L/usr/local/cuda-7.0/lib -L/usr/local/cuda-7.0/lib64 -lcuda -lcudart -lboost_python -lpython2.7 -lpthread -ldl -lutil -o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/codepy.temp.46c837d5cb878372cdb2886923bd37a6.7ceed7a1666bbc7019867d851e0660de.module.so\nUsing the config file specified: \' Data/Model/DiaCfg/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.cfg \'\nSpeech file spnsp_file not specified, continuing without it...\n...Reading in HTK feature file...\nINFO: total number of frames read:  779339\n...Reading in speech/nonspeech file...\n ====================== CLUSTERING ON SUBSET ====================== \nGMM SPECIALIZER: USING CUDA\nrecompiling for non-existent cache dir (/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/38d44195c321f1de7e40420e316a4733).\ng++ -pthread -fno-strict-aliasing -g -O2 -g -fwrapv -O3 -Wall -fPIC -c -DNDEBUG -I/usr/local/python/2.7.8/include/python2.7 -I/home/hxx124/myPython/virtualenv-1.9/ENV/lib/python2.7/site-packages/numpy/core/include -I/usr/local/cuda-7.0/include /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/38d44195c321f1de7e40420e316a4733/module.cpp -o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/38d44195c321f1de7e40420e316a4733/module.o\nrecompiling for non-existent cache dir (/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4).\nnvcc -Xcompiler -pthread,-fno-strict-aliasing,-g,-O2,-g,-fwrapv,-O3,-Wall,-Wstrict-prototypes,-fPIC -shared -Xcompiler -fPIC -arch=sm_20 -c -DNDEBUG -U__BLOCKS__ -I/usr/local/python/2.7.8/include/python2.7 -I/home/hxx124/myPython/virtualenv-1.9/ENV/lib/python2.7/site-packages/numpy/core/include -I/usr/local/cuda-7.0/include /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu -o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.o\ncc1plus: warning: command line option "-Wstrict-prototypes" is valid for Ada/C/ObjC but not for C++\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(193): warning: variable "maxsize" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(194): warning: variable "n" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(195): warning: variable "num_threads" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(290): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(290): warning: variable "col" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(378): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(378): warning: variable "col" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(430): warning: variable "num_threads" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(494): warning: variable "num_components" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(756): warning: variable "avgvar" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1230): warning: variable "num_threads" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1231): warning: variable "num_dimensions" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1232): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1232): warning: variable "col" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1372): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1372): warning: variable "col" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1460): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1460): warning: variable "col" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1512): warning: variable "num_threads" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1576): warning: variable "num_components" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1839): warning: variable "avgvar" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(25): warning: function "ToFixedPoint" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(31): warning: function "ToFloatPoint" was declared but never referenced\n\ncc1plus: warning: command line option "-Wstrict-prototypes" is valid for Ada/C/ObjC but not for C++\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(193): warning: variable "maxsize" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(194): warning: variable "n" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(195): warning: variable "num_threads" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(290): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(290): warning: variable "col" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(378): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(378): warning: variable "col" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(430): warning: variable "num_threads" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(494): warning: variable "num_components" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(756): warning: variable "avgvar" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1230): warning: variable "num_threads" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1231): warning: variable "num_dimensions" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1232): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1232): warning: variable "col" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1372): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1372): warning: variable "col" was declared b/usr/bin/ld: skipping incompatible /usr/local/cuda-7.0/lib/libcudart.so when searching for -lcudart\nut never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1460): warning: variable "row" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1460): warning: variable "col" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1512): warning: variable "num_threads" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1576): warning: variable "num_components" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(1839): warning: variable "avgvar" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(25): warning: function "ToFixedPoint" was declared but never referenced\n\n/tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.cu(31): warning: function "ToFloatPoint" was declared but never referenced\n\ncc1plus: warning: command line option "-Wstrict-prototypes" is valid for Ada/C/ObjC but not for C++\ncc1plus: warning: command line option "-Wstrict-prototypes" is valid for Ada/C/ObjC but not for C++\n\n\ng++ -pthread -fno-strict-aliasing -g -O2 -g -fwrapv -O3 -Wall -fPIC -pthread -shared -Xlinker -export-dynamic -DNDEBUG -I/usr/local/python/2.7.8/include/python2.7 -I/home/hxx124/myPython/virtualenv-1.9/ENV/lib/python2.7/site-packages/numpy/core/include -I/usr/local/cuda-7.0/include /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/38d44195c321f1de7e40420e316a4733/module.o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/11dee452985a0f3b9b6a40453d090df4/gpu.o -L/usr/local/python/2.7.8/lib -L/usr/local/cuda-7.0/lib -L/usr/local/cuda-7.0/lib64 -lcuda -lcudart -lboost_python -lpython2.7 -lpthread -ldl -lutil -o /tmp/pbsjobs.460477.hpc/asp_cache_hxx124/codepy.temp.38d44195c321f1de7e40420e316a4733.11dee452985a0f3b9b6a40453d090df4.module.so\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 15]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 15, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 15, 10, 10]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 5, 10, 10, 15, 10, 15]\n size of each cluster: [5, 5, 5, 5, 5, 5, 5, 10, 10, 15, 10, 20]\n size of each cluster: [5, 5, 5, 5, 5, 5, 10, 15, 10, 20, 15]\n size of each cluster: [5, 5, 5, 5, 5, 5, 10, 15, 10, 35]\n size of each cluster: [5, 5, 5, 5, 5, 15, 10, 35, 15]\n size of each cluster: [5, 5, 5, 5, 10, 35, 15, 20]\n size of each cluster: [5, 5, 5, 35, 15, 20, 15]\n size of each cluster: [5, 5, 35, 15, 20, 20]\n size of each cluster: [5, 5, 15, 20, 55]\n size of each cluster: [5, 15, 20, 60]\n size of each cluster: [5, 60, 35]\n size of each cluster: [35, 65]\n size of each cluster: [100]\n=== Total clustering time:  956.094987154\n=== Final size of each cluster: [100]\n...Writing out RTTM file...\nDONE writing RTTM file')
(0, '')

In [19]:

    

# Below you can find an example on training a fresh model
# However, please make sure that the sample rate of your training data is the same with your testing data
model_gender_dir = "Data/Model/Gender/"
model_gender_nm = 'gender.model'
Gender = SR.GMMRec() # Create a new recognizer
female_fn = model_gender_dir+'female.wav' # choose the training file for female
male_fn = model_gender_dir+'male.wav' # choose the training file for male
Gender.enroll_file('Female', female_fn) # enroll the female audio 
Gender.enroll_file('Male', male_fn) # enroll the male audio 
Gender.train() # train the GMMs after you enroll all the training data 
Gender.dump(model_gender_dir+model_gender_nm) # save the trained model into a file named "gender.model" for future use


    

In [7]:

    

# Gender Identification based on Speaker Diarization
model_gender_dir = "Data/Model/Gender/"
model_gender_nm = 'gender.model'
model_gender = model_gender_dir+model_gender_nm # Read from the model
gen_dir = "Data/Gender/"
commands.getstatusoutput("mkdir -p "+gen_dir)
ext_gen = ".gen"
dia2spk(model_gender, audio, dia, meta, gen_dir, ext_gen)
gen = os.path.join(gen_dir, name+ext_gen)


    

    




(0, '')

In [8]:

    

# Gender Identification without Speaker Diarization
model_genr = model_gender_dir+model_gender_nm
genr_dir = "Data/Gender/"
commands.getstatusoutput("mkdir -p "+genr_dir)
ext_genr = ".genr"
spk(model_genr, audio, meta, genr_dir, ext_genr)
genr = os.path.join(genr_dir, name+ext_genr)


    

    




(0, '')

In [9]:

    

# you can train a fresh model as follows:

model_speaker_dir = "Data/Model/Speaker/" # The training data should already be here!! 
model_speaker_nm = 'speaker.model'
Speaker = SR.GMMRec() # Create a new recognizer
other_fn = model_speaker_dir+'Imposter.wav' # choose the training file for female
trump_fn = model_speaker_dir+'Trump.wav' # choose the training file for male
Speaker.enroll_file('Other', other_fn) # enroll the female audio 
Speaker.enroll_file('Trump', trump_fn) # enroll the male audio 
# You can add more speakers here following the syntax above
Speaker.train() # train the GMMs after you enroll all the training data 
Speaker.dump(model_speaker_dir+model_speaker_nm) # save the trained model into a file named "gender.model" for future use


    

In [11]:

    

# Speaker Recognition based Speaker Diarization
model_spk = model_speaker_dir+model_speaker_nm
spk_dir = "Data/Speaker/"
commands.getstatusoutput("mkdir -p "+spk_dir)
ext_spk = ".spk"
spk(model_spk, audio, meta, spk_dir, ext_spk)
spk = os.path.join(spk_dir, name+ext_spk)


    

    




(0, '')

In [10]:

    

# Speaker Recognition without Speaker Diarization
model_speaker_dir = "Data/Model/Speaker/"
model_speaker_nm = 'speaker.model'
model_spkr = model_speaker_dir+model_speaker_nm
spkr_dir = "Data/Speaker/"
commands.getstatusoutput("mkdir -p "+spkr_dir)
ext_spkr = ".spkr"
spk(model_spkr, audio, meta, spkr_dir, ext_spkr)
spkr = os.path.join(spkr_dir, name+ext_spkr)


    

    




(0, '')

In [31]:

    

commands.getstatusoutput("ffmpeg -i "+video+" -vn -f wav -ab 44100 "+
                             os.path.join(audio_dir,video.split("/")[-1].split(".")[0]+audio_ext))


    

    
Out[31]:





(0,
 "ffmpeg version 2.8.2 Copyright (c) 2000-2015 the FFmpeg developers\n  built with gcc 4.9.3 (GCC)\n  configuration: --prefix=/usr/local/ffmpeg/2.8.2 --enable-shared\n  libavutil      54. 31.100 / 54. 31.100\n  libavcodec     56. 60.100 / 56. 60.100\n  libavformat    56. 40.101 / 56. 40.101\n  libavdevice    56.  4.100 / 56.  4.100\n  libavfilter     5. 40.101 /  5. 40.101\n  libswscale      3.  1.101 /  3.  1.101\n  libswresample   1.  2.101 /  1.  2.101\nInput #0, mov,mp4,m4a,3gp,3g2,mj2, from 'Data/Video/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.mp4':\n  Metadata:\n    major_brand     : isom\n    minor_version   : 512\n    compatible_brands: isomiso2avc1mp41\n    encoder         : Lavf56.15.102\n  Duration: 02:09:53.49, start: 0.046440, bitrate: 554 kb/s\n    Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 640x362 [SAR 181:180 DAR 16:9], 449 kb/s, 29.97 fps, 29.97 tbr, 11988 tbn, 59.94 tbc (default)\n    Metadata:\n      handler_name    : VideoHandler\n    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 96 kb/s (default)\n    Metadata:\n      handler_name    : SoundHandler\nOutput #0, wav, to 'Data/Audio/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.wav':\n  Metadata:\n    major_brand     : isom\n    minor_version   : 512\n    compatible_brands: isomiso2avc1mp41\n    ISFT            : Lavf56.40.101\n    Stream #0:0(und): Audio: pcm_s16le ([1][0][0][0] / 0x0001), 44100 Hz, stereo, s16, 1411 kb/s (default)\n    Metadata:\n      handler_name    : SoundHandler\n      encoder         : Lavc56.60.100 pcm_s16le\nStream mapping:\n  Stream #0:1 -> #0:0 (aac (native) -> pcm_s16le (native))\nPress [q] to stop, [?] for help\nsize=   29208kB time=00:02:49.55 bitrate=1411.2kbits/s    \rsize=   65540kB time=00:06:20.45 bitrate=1411.2kbits/s    \rsize=  103432kB time=00:10:00.42 bitrate=1411.2kbits/s    \rsize=  134116kB time=00:12:58.54 bitrate=1411.2kbits/s    \rsize=  172416kB time=00:16:40.87 bitrate=1411.2kbits/s    \rsize=  199840kB time=00:19:20.06 bitrate=1411.2kbits/s    \rsize=  237368kB time=00:22:57.91 bitrate=1411.2kbits/s    \rsize=  267728kB time=00:25:54.15 bitrate=1411.2kbits/s    \rsize=  305184kB time=00:29:31.58 bitrate=1411.2kbits/s    \rsize=  333776kB time=00:32:17.56 bitrate=1411.2kbits/s    \rsize=  370736kB time=00:35:52.11 bitrate=1411.2kbits/s    \rsize=  399684kB time=00:38:40.16 bitrate=1411.2kbits/s    \rsize=  435832kB time=00:42:09.99 bitrate=1411.2kbits/s    \rsize=  465496kB time=00:45:02.19 bitrate=1411.2kbits/s    \rsize=  502972kB time=00:48:39.74 bitrate=1411.2kbits/s    \rsize=  532612kB time=00:51:31.80 bitrate=1411.2kbits/s    \rsize=  569576kB time=00:55:06.38 bitrate=1411.2kbits/s    \rsize=  599108kB time=00:57:57.81 bitrate=1411.2kbits/s    \rsize=  636200kB time=01:01:33.13 bitrate=1411.2kbits/s    \rsize=  665628kB time=01:04:23.96 bitrate=1411.2kbits/s    \rsize=  703164kB time=01:08:01.85 bitrate=1411.2kbits/s    \rsize=  731736kB time=01:10:47.71 bitrate=1411.2kbits/s    \rsize=  768276kB time=01:14:19.83 bitrate=1411.2kbits/s    \rsize=  796296kB time=01:17:02.48 bitrate=1411.2kbits/s    \rsize=  833360kB time=01:20:37.64 bitrate=1411.2kbits/s    \rsize=  863068kB time=01:23:30.09 bitrate=1411.2kbits/s    \rsize=  900060kB time=01:27:04.83 bitrate=1411.2kbits/s    \rsize=  930248kB time=01:30:00.07 bitrate=1411.2kbits/s    \rsize=  967160kB time=01:33:34.35 bitrate=1411.2kbits/s    \rsize=  995508kB time=01:36:18.91 bitrate=1411.2kbits/s    \rsize= 1031964kB time=01:39:50.53 bitrate=1411.2kbits/s    \rsize= 1061324kB time=01:42:40.97 bitrate=1411.2kbits/s    \rsize= 1099444kB time=01:46:22.25 bitrate=1411.2kbits/s    \rsize= 1128456kB time=01:49:10.67 bitrate=1411.2kbits/s    \rsize= 1165560kB time=01:52:46.06 bitrate=1411.2kbits/s    \rsize= 1195448kB time=01:55:39.56 bitrate=1411.2kbits/s    \rsize= 1232888kB time=01:59:16.90 bitrate=1411.2kbits/s    \rsize= 1262364kB time=02:02:08.00 bitrate=1411.2kbits/s    \rsize= 1299544kB time=02:05:43.83 bitrate=1411.2kbits/s    \rsize= 1328524kB time=02:08:32.06 bitrate=1411.2kbits/s    \rsize= 1342536kB time=02:09:53.40 bitrate=1411.2kbits/s    \nvideo:0kB audio:1342536kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.000006%")

In [6]:

    

wav.read(audio)


    

    
Out[6]:





(44100, array([[0, 0],
        [0, 0],
        [0, 0],
        ..., 
        [0, 0],
        [0, 0],
        [0, 0]], dtype=int16))

In [8]:

    

audio


    

    
Out[8]:





'Data/Audio/2015-08-07_0050_US_FOX-News_US_Presidential_Politics.wav'

In [ ]: