Class: Diarize::Speaker

Inherits:
Object
  • Object
show all
Includes:
JBLAS, ToRdf
Defined in:
lib/diarize/speaker.rb

Constant Summary collapse

@@log_likelihood_threshold = 
-33
@@detection_threshold = 
0.2
@@speakers = 
{}

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(uri = nil, gender = nil, model_file = nil) ⇒ Speaker

Returns a new instance of Speaker.



35
36
37
38
39
40
 
# File 'lib/diarize/speaker.rb', line 35

def initialize(uri = nil, gender = nil, model_file = nil)
  @model = Speaker.load_model(model_file) if model_file
  @uri = uri
  @gender = gender
  @normalized = false
end

Instance Attribute Details

#gender ⇒ Object (readonly)

Returns the value of attribute gender.



33
34
35
 
# File 'lib/diarize/speaker.rb', line 33

def gender
  @gender
end

#model ⇒ Object

Returns the value of attribute model.



32
33
34
 
# File 'lib/diarize/speaker.rb', line 32

def model
  @model
end

#model_uri ⇒ Object

Returns the value of attribute model_uri.



32
33
34
 
# File 'lib/diarize/speaker.rb', line 32

def model_uri
  @model_uri
end

#normalized ⇒ Object

Returns the value of attribute normalized.



32
33
34
 
# File 'lib/diarize/speaker.rb', line 32

def normalized
  @normalized
end

Class Method Details

.detection_threshold ⇒ Object 



66
67
68
 
# File 'lib/diarize/speaker.rb', line 66

def self.detection_threshold
  @@detection_threshold
end

.detection_threshold=(threshold) ⇒ Object 



62
63
64
 
# File 'lib/diarize/speaker.rb', line 62

def self.detection_threshold=(threshold)
  @@detection_threshold = threshold
end

.divergence(speaker1, speaker2) ⇒ Object 



79
80
81
82
83
84
85
 
# File 'lib/diarize/speaker.rb', line 79

def self.divergence(speaker1, speaker2)
  # TODO bundle in mean_log_likelihood to weight down unlikely models? 
  return unless speaker1.model and speaker2.model
  # MAP Gaussian divergence
  # See "A model space framework for efficient speaker detection", Interspeech'05
  divergence_lium(speaker1, speaker2)
end

.divergence_lium(speaker1, speaker2) ⇒ Object 



87
88
89
 
# File 'lib/diarize/speaker.rb', line 87

def self.divergence_lium(speaker1, speaker2)
  fr.lium.spkDiarization.libModel.Distance.GDMAP(speaker1.model, speaker2.model)
end

.divergence_ruby(speaker1, speaker2) ⇒ Object 



91
92
93
 
# File 'lib/diarize/speaker.rb', line 91

def self.divergence_ruby(speaker1, speaker2)
  SuperVector.divergence(speaker1.supervector, speaker2.supervector)
end

.find_or_create(uri, gender) ⇒ Object 



74
75
76
77
 
# File 'lib/diarize/speaker.rb', line 74

def self.find_or_create(uri, gender)
  return @@speakers[uri] if @@speakers[uri]
  @@speakers[uri] = Speaker.new(uri, gender)
end

.load_model(filename) ⇒ Object 



70
71
72
 
# File 'lib/diarize/speaker.rb', line 70

def self.load_model(filename)
  read_gmm(filename)
end

.match(speakers) ⇒ Object 



105
106
107
 
# File 'lib/diarize/speaker.rb', line 105

def self.match(speakers)
  speakers.combination(2).select { |s1, s2| s1.same_speaker_as(s2) }
end

.match_sets(speakers1, speakers2) ⇒ Object 



95
96
97
98
99
100
101
102
103
 
# File 'lib/diarize/speaker.rb', line 95

def self.match_sets(speakers1, speakers2)
  matches = []
  speakers1.each do |s1|
    speakers2.each do |s2|
      matches << [ s1, s2 ] if s1.same_speaker_as(s2)
    end
  end
  matches
end

.ubm ⇒ Object 



42
43
44
45
46
47
 
# File 'lib/diarize/speaker.rb', line 42

def self.ubm
  speaker = Speaker.new
  speaker.normalized = true
  speaker.model = Speaker.load_model(File.join(File.expand_path(File.dirname(__FILE__)), 'ubm.gmm'))
  speaker
end

Instance Method Details

#mean_log_likelihood ⇒ Object 



49
50
51
 
# File 'lib/diarize/speaker.rb', line 49

def mean_log_likelihood
  @mean_log_likelihood ? @mean_log_likelihood : model.mean_log_likelihood # Will be NaN if model was loaded from somewhere
end

#mean_log_likelihood=(mll) ⇒ Object 



53
54
55
 
# File 'lib/diarize/speaker.rb', line 53

def mean_log_likelihood=(mll)
  @mean_log_likelihood = mll
end

#namespaces ⇒ Object 



145
146
147
 
# File 'lib/diarize/speaker.rb', line 145

def namespaces
  super.merge 'ws' => 'http://wsarchive.prototype0.net/ontology/'
end

#normalize! ⇒ Object 



109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
 
# File 'lib/diarize/speaker.rb', line 109

def normalize!
  unless @normalized
    # Applies M-Norm from "D-MAP: a Distance-Normalized MAP Estimation of Speaker Models for Automatic Speaker Verification"
    # to the associated GMM, placing it on a unit hyper-sphere with a UBM centre (model will be at distance one from the UBM
    # according to GDMAP)
    # Using supervectors: vector = (1.0 / distance_to_ubm) * vector + (1.0 - 1.0 / distance_to_ubm) * ubm_vector
    speaker_ubm = Speaker.ubm
    distance_to_ubm = Math.sqrt(Speaker.divergence(self, speaker_ubm))
    model.nb_of_components.times do |k|
      gaussian = model.components.get(k)
      gaussian.dim.times do |i|
        normalized_mean = (1.0 / distance_to_ubm) * gaussian.mean(i) + (1.0 - 1.0 / distance_to_ubm)  * speaker_ubm.model.components.get(k).mean(i)
        gaussian.set_mean(i, normalized_mean) 
      end
    end
    @normalized = true
  end
  @normalized
end

#rdf_mapping ⇒ Object 



157
158
159
 
# File 'lib/diarize/speaker.rb', line 157

def rdf_mapping
  { 'ws:gender' => gender, 'ws:model' => model_uri, 'ws:mean_log_likelihood' => model.mean_log_likelihood, 'ws:supervector_hash' => supervector.hash.to_s }
end

#same_speaker_as(other) ⇒ Object 



129
130
131
132
133
134
135
136
 
# File 'lib/diarize/speaker.rb', line 129

def same_speaker_as(other)
  # Detection score defined in Ben2005
  return unless [ self.mean_log_likelihood, other.mean_log_likelihood ].min > @@log_likelihood_threshold
  self.normalize!
  other.normalize!
  detection_score = 1.0 - Speaker.divergence(other, self)
  detection_score > @@detection_threshold
end

#save_model(filename) ⇒ Object 



57
58
59
60
 
# File 'lib/diarize/speaker.rb', line 57

def save_model(filename)
  # TODO perhaps a warning if a normalised model is being saved?
  write_gmm(filename, @model)
end

#supervector ⇒ Object 



138
139
140
141
 
# File 'lib/diarize/speaker.rb', line 138

def supervector
  # TODO: cache only when normalized
  @supervector ||= SuperVector.generate_from_model(model)
end

#type_uri ⇒ Object 



153
154
155
 
# File 'lib/diarize/speaker.rb', line 153

def type_uri
  'ws:Speaker'
end

#uri ⇒ Object 



149
150
151
 
# File 'lib/diarize/speaker.rb', line 149

def uri
  @uri
end