diarize-ruby 0.3.7 → 0.3.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: acabfb7160ffeb0ee59815441c7108aebd3dbc57
-  data.tar.gz: 1e1ac96bdeb06c3546ef652f1b8324f38c42316f
+  metadata.gz: a0ca7dc41b19eb2fe4496e20cb87f260ebf52141
+  data.tar.gz: 85d3e9cda2554f69737ffb11fee7ff0b5e9b54e9
 SHA512:
-  metadata.gz: f453a319e2d7e4084fa95fa33240a78773bcf6d7dddf8551b02337a5a9ed294ebc48317c52ba2d845fb2f095f86397038e0759124aa4409721a1f3d5cf2f7786
-  data.tar.gz: 6be77004bd7ba178fba11f5b3e8510357f7c2708b5fc49bfc302a570af8b6e0727603cd776373e157c1227e7e354b1abf3bec63810d87f5df7b72a7148e99dbc
+  metadata.gz: 219f481f7243ac904a35af383530b0308ddb0476afae76efcd954ac444112f9509080beabbd6797ff3e33266f48e75cce22af25385cf9170572d35c73d9feb61
+  data.tar.gz: 844543077abb226237e1e6ccdb28a3564a977360361686a820b00e190dd243f0545f8ee8313f25effaf192a4622e543482da561d78ade9567d331ad633515397

data/CHANGELOG.md

@@ -1,3 +1,11 @@
+## [v0.3.8] - 2016-11-29
+
+- Speaker#load_model to return Java object instead of Rjb wrapper
+- Added sha to SuperVector
+- Added `as_json`/`to_json` for speakers
+- `to_rdf`, `as_json` and `to_json` to return supervector_sha for speaker identification
+- Refactored Rjb wrapper into separate folder
+
 ## [v0.3.7] - 2016-11-18
 
 - Sort segments by start time by default

data/lib/diarize.rb

@@ -9,9 +9,11 @@ require "matrix"
 
 require "uri"
 require "open-uri"
-require "digest/md5"
+require "digest"
 require "to_rdf"
 
+require "rjb/java_object_wrapper"
+
 require "diarize/version"
 require "diarize/audio"
 require "diarize/segment"
@@ -19,105 +21,3 @@ require "diarize/speaker"
 require "diarize/segmentation"
 require "diarize/super_vector"
 require "diarize/server"
-
-# Extenions to the {Ruby-Java Bridge}[http://rjb.rubyforge.org/] module that
-# adds a generic Java object wrapper class.
-module Rjb
-  # A generic wrapper for a Java object loaded via the Ruby Java Bridge.  The
-  # wrapper class handles intialization and stringification, and passes other
-  # method calls down to the underlying Java object.  Objects returned by the
-  # underlying Java object are converted to the appropriate Ruby object.
-  #
-  # This object is enumerable, yielding items in the order defined by the Java
-  # object's iterator.
-  class JavaObjectWrapper
-    include Enumerable
-
-    # The underlying Java object.
-    attr_reader :java_object
-
-    # Initialize with a Java object <em>obj</em>.  If <em>obj</em> is a
-    # String, assume it is a Java class name and instantiate it.  Otherwise,
-    # treat <em>obj</em> as an instance of a Java object.
-    def initialize(obj, *args)
-      @java_object = obj.class == String ?
-      Rjb::import(obj).send(:new, *args) : obj
-    end
-
-    # Enumerate all the items in the object using its iterator.  If the object
-    # has no iterator, this function yields nothing.
-    def each
-      if @java_object.getClass.getMethods.any? {|m| m.getName == "iterator"}
-        i = @java_object.iterator
-        while i.hasNext
-          yield wrap_java_object(i.next)
-        end
-      end
-    end # each
-
-    # Reflect unhandled method calls to the underlying Java object.
-    def method_missing(m, *args)
-      wrap_java_object(@java_object.send(m, *args))
-    end
-
-    # Convert a value returned by a call to the underlying Java object to the
-    # appropriate Ruby object as follows:
-    # * RJB objects are placed inside a generic JavaObjectWrapper wrapper.
-    # * <tt>java.util.ArrayList</tt> objects are converted to Ruby Arrays.
-    # * <tt>java.util.HashSet</tt> objects are converted to Ruby Sets
-    # * Other objects are left unchanged.
-    #
-    # This function is applied recursively to items in collection objects such
-    # as set and arrays.
-    def wrap_java_object(object)
-      if object.kind_of?(Array)
-        object.collect {|item| wrap_java_object(item)}
-      # Ruby-Java Bridge Java objects all have a _classname member which tells
-      # the name of their Java class.
-      elsif object.respond_to?(:_classname)
-        case object._classname
-        when /java\.util\.ArrayList/
-          # Convert java.util.ArrayList objects to Ruby arrays.
-          array_list = []
-          object.size.times do
-            |i| array_list << wrap_java_object(object.get(i))
-          end
-          array_list
-        when /java\.util\.HashSet/
-          # Convert java.util.HashSet objects to Ruby sets.
-          set = Set.new
-          i = object.iterator
-          while i.hasNext
-            set << wrap_java_object(i.next)
-          end
-          set
-        else
-          # Pass other RJB objects off to a handler.
-          wrap_rjb_object(object)
-        end # case
-      else
-        # Return non-RJB objects unchanged.
-        object
-      end # if
-    end # wrap_java_object
-
-    # By default, all RJB classes other than <tt>java.util.ArrayList</tt> and
-    # <tt>java.util.HashSet</tt> go in a generic wrapper.  Derived classes may
-    # change this behavior.
-    def wrap_rjb_object(object)
-      JavaObjectWrapper.new(object)
-    end
-
-    # Show the classname of the underlying Java object.
-    def inspect
-      "<#{@java_object._classname}>"
-    end
-
-    # Use the underlying Java object's stringification.
-    def to_s
-      toString
-    end
-
-    protected :wrap_java_object, :wrap_rjb_object
-  end # JavaObjectWrapper
-end # Rjb

data/lib/diarize/audio.rb

@@ -155,7 +155,7 @@ module Diarize
 
       # Populating the speakers with their GMMs
       gmm_vect.each_with_index do |speaker_model, i|
-        speakers[i].model = speaker_model
+        speakers[i].model = speaker_model.java_object
       end
     end
 

data/lib/diarize/speaker.rb

@@ -1,90 +1,110 @@
 module Diarize
   class Speaker
+    include ToRdf
+
     @@log_likelihood_threshold = -33
     @@detection_threshold      = 0.2
+    @@speakers                 = {}
 
-    @@speakers = {}
-
-    attr_accessor :model_uri, :model, :normalized
+    attr_accessor :model_uri, :model
     attr_reader :gender
+    attr_writer :normalized
 
     def initialize(uri = nil, gender = nil, model_file = nil)
-      @model = Speaker.load_model(model_file) if model_file
-      @uri = uri
-      @gender = gender
+      @model      = Speaker.load_model(model_file) if model_file
+      @uri        = uri
+      @gender     = gender
       @normalized = false
     end
 
-    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
+    class << self
 
-    def mean_log_likelihood
-      @mean_log_likelihood ? @mean_log_likelihood : model.mean_log_likelihood # Will be NaN if model was loaded from somewhere
-    end
+      def ubm
+        speaker = Speaker.new
+        speaker.normalized = true
+        speaker.model = Speaker.load_model(File.join(File.expand_path(File.dirname(__FILE__)), 'ubm.gmm'))
+        speaker
+      end
 
-    def mean_log_likelihood=(mll)
-      @mean_log_likelihood = mll
-    end
+      def detection_threshold=(threshold)
+        @@detection_threshold = threshold
+      end
 
-    def save_model(filename)
-      # TODO perhaps a warning if a normalised model is being saved?
-      write_gmm(filename, @model)
-    end
+      def detection_threshold
+        @@detection_threshold
+      end
 
-    def self.detection_threshold=(threshold)
-      @@detection_threshold = threshold
-    end
+      def load_model(filename)
+        read_gmm(filename)
+      end
 
-    def self.detection_threshold
-      @@detection_threshold
-    end
+      def find_or_create(uri, gender)
+        return @@speakers[uri] if @@speakers[uri]
+        @@speakers[uri] = Speaker.new(uri, gender)
+      end
 
-    def self.load_model(filename)
-      read_gmm(filename)
-    end
+      def 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
 
-    def self.find_or_create(uri, gender)
-      return @@speakers[uri] if @@speakers[uri]
-      @@speakers[uri] = Speaker.new(uri, gender)
-    end
+      def divergence_lium(speaker1, speaker2)
+        Rjb::import('fr.lium.spkDiarization.libModel.Distance').GDMAP(speaker1.model, speaker2.model)
+      end
 
-    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
+      def divergence_ruby(speaker1, speaker2)
+        SuperVector.divergence(speaker1.supervector, speaker2.supervector)
+      end
 
-    def self.divergence_lium(speaker1, speaker2)
-      # fr.lium.spkDiarization.libModel.Distance.GDMAP(speaker1.model, speaker2.model)
-      Rjb::import('fr.lium.spkDiarization.libModel.Distance').GDMAP(speaker1.model, speaker2.model)
+      def 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
+
+      def match(speakers)
+        speakers.combination(2).select { |s1, s2| s1.same_speaker_as(s2) }
+      end
+
+      protected
+
+      def read_gmm(filename)
+        gmmlist = Rjb::JavaObjectWrapper.new("java.util.ArrayList")
+        input = Rjb::import('fr.lium.spkDiarization.lib.IOFile').new(filename, 'rb')
+        input.open
+        Rjb::import('fr.lium.spkDiarization.libModel.ModelIO').readerGMMContainer(input, gmmlist.java_object)
+        input.close
+        gmmlist.to_a.first.java_object
+      end
+
+    end # class
+
+    def mean_log_likelihood
+      @mean_log_likelihood ? @mean_log_likelihood : model.mean_log_likelihood # Will be NaN if model was loaded from somewhere
     end
 
-    def self.divergence_ruby(speaker1, speaker2)
-      SuperVector.divergence(speaker1.supervector, speaker2.supervector)
+    def mean_log_likelihood=(mll)
+      @mean_log_likelihood = mll
     end
 
-    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
+    def save_model(filename)
+      # TODO perhaps a warning if a normalized model is being saved?
+      write_gmm(filename, @model)
     end
 
-    def self.match(speakers)
-      speakers.combination(2).select { |s1, s2| s1.same_speaker_as(s2) }
+    def normalized?
+      !!@normalized
     end
 
     def normalize!
-      unless @normalized
+      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)
@@ -113,12 +133,15 @@ module Diarize
     end
 
     def supervector
-      # TODO: cache only when normalized
-      @supervector ||= SuperVector.generate_from_model(model)
+      if normalized?
+        @supervector ||= begin
+          SuperVector.generate_from_model(model)
+        end
+      else
+        SuperVector.generate_from_model(model)
+      end
     end
 
-    include ToRdf
-
     def namespaces
       super.merge 'ws' => 'http://wsarchive.prototype0.net/ontology/'
     end
@@ -132,23 +155,31 @@ module Diarize
     end
 
     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 }
+      {
+        'ws:gender' => gender,
+        'ws:model' => model_uri,
+        'ws:mean_log_likelihood' => mean_log_likelihood,
+        'ws:supervector_hash' => supervector.hash.to_s,
+        'ws:supervector_sha' => supervector.sha
+      }
     end
 
-    protected
+    def as_json
+      {
+        'gender' => gender,
+        'model' => model_uri,
+        'mean_log_likelihood' => mean_log_likelihood,
+        'supervector_hash' => supervector.hash.to_s,
+        'supervector_sha' => supervector.sha
+      }
+    end
 
-    def self.read_gmm(filename)
-      # gmmlist = java.util.ArrayList.new
-      gmmlist = Rjb::JavaObjectWrapper.new("java.util.ArrayList")
-      # input = fr.lium.spkDiarization.lib.IOFile.new(filename, 'rb')
-      input = Rjb::import('fr.lium.spkDiarization.lib.IOFile').new(filename, 'rb')
-      input.open
-      # fr.lium.spkDiarization.libModel.ModelIO.readerGMMContainer(input, gmmlist)
-      Rjb::import('fr.lium.spkDiarization.libModel.ModelIO').readerGMMContainer(input, gmmlist.java_object)
-      input.close
-      gmmlist.to_a.first.java_object
+    def to_json
+      as_json.to_json
     end
 
+    protected
+
     def write_gmm(filename, model)
       # gmmlist = java.util.ArrayList.new
       gmmlist = Rjb::JavaObjectWrapper.new("java.util.ArrayList")
@@ -160,5 +191,6 @@ module Diarize
       Rjb::import('fr.lium.spkDiarization.libModel.ModelIO').writerGMMContainer(output, gmmlist.java_object)
       output.close
     end
+
   end # Speaker
 end

data/lib/diarize/super_vector.rb

@@ -71,5 +71,13 @@ module Diarize
       @vector.hash
     end
 
+    def sha
+      Digest::SHA256.hexdigest(hash.to_s)
+    end
+
+    def to_a
+      @vector.to_a
+    end
+
   end # SuperVector
 end

data/lib/diarize/version.rb

@@ -1,3 +1,3 @@
 module Diarize
-  VERSION = "0.3.7"
+  VERSION = "0.3.8"
 end

data/lib/rjb/java_object_wrapper.rb

@@ -0,0 +1,101 @@
+# Extenions to the {Ruby-Java Bridge}[http://rjb.rubyforge.org/] module that
+# adds a generic Java object wrapper class.
+module Rjb
+  # A generic wrapper for a Java object loaded via the Ruby Java Bridge.  The
+  # wrapper class handles intialization and stringification, and passes other
+  # method calls down to the underlying Java object.  Objects returned by the
+  # underlying Java object are converted to the appropriate Ruby object.
+  #
+  # This object is enumerable, yielding items in the order defined by the Java
+  # object's iterator.
+  class JavaObjectWrapper
+    include Enumerable
+
+    # The underlying Java object.
+    attr_reader :java_object
+
+    # Initialize with a Java object <em>obj</em>.  If <em>obj</em> is a
+    # String, assume it is a Java class name and instantiate it.  Otherwise,
+    # treat <em>obj</em> as an instance of a Java object.
+    def initialize(obj, *args)
+      @java_object = obj.class == String ?
+      Rjb::import(obj).send(:new, *args) : obj
+    end
+
+    # Enumerate all the items in the object using its iterator.  If the object
+    # has no iterator, this function yields nothing.
+    def each
+      if @java_object.getClass.getMethods.any? {|m| m.getName == "iterator"}
+        i = @java_object.iterator
+        while i.hasNext
+          yield wrap_java_object(i.next)
+        end
+      end
+    end # each
+
+    # Reflect unhandled method calls to the underlying Java object.
+    def method_missing(m, *args)
+      wrap_java_object(@java_object.send(m, *args))
+    end
+
+    # Convert a value returned by a call to the underlying Java object to the
+    # appropriate Ruby object as follows:
+    # * RJB objects are placed inside a generic JavaObjectWrapper wrapper.
+    # * <tt>java.util.ArrayList</tt> objects are converted to Ruby Arrays.
+    # * <tt>java.util.HashSet</tt> objects are converted to Ruby Sets
+    # * Other objects are left unchanged.
+    #
+    # This function is applied recursively to items in collection objects such
+    # as set and arrays.
+    def wrap_java_object(object)
+      if object.kind_of?(Array)
+        object.collect {|item| wrap_java_object(item)}
+      # Ruby-Java Bridge Java objects all have a _classname member which tells
+      # the name of their Java class.
+      elsif object.respond_to?(:_classname)
+        case object._classname
+        when /java\.util\.ArrayList/
+          # Convert java.util.ArrayList objects to Ruby arrays.
+          array_list = []
+          object.size.times do
+            |i| array_list << wrap_java_object(object.get(i))
+          end
+          array_list
+        when /java\.util\.HashSet/
+          # Convert java.util.HashSet objects to Ruby sets.
+          set = Set.new
+          i = object.iterator
+          while i.hasNext
+            set << wrap_java_object(i.next)
+          end
+          set
+        else
+          # Pass other RJB objects off to a handler.
+          wrap_rjb_object(object)
+        end # case
+      else
+        # Return non-RJB objects unchanged.
+        object
+      end # if
+    end # wrap_java_object
+
+    # By default, all RJB classes other than <tt>java.util.ArrayList</tt> and
+    # <tt>java.util.HashSet</tt> go in a generic wrapper.  Derived classes may
+    # change this behavior.
+    def wrap_rjb_object(object)
+      JavaObjectWrapper.new(object)
+    end
+
+    # Show the classname of the underlying Java object.
+    def inspect
+      "<#{@java_object._classname}>"
+    end
+
+    # Use the underlying Java object's stringification.
+    def to_s
+      toString
+    end
+
+    protected :wrap_java_object, :wrap_rjb_object
+  end # JavaObjectWrapper
+end # Rjb

data/test/speaker_test.rb

@@ -2,6 +2,9 @@ require 'test_helper'
 require 'tempfile'
 
 class SpeakerTest < Test::Unit::TestCase
+  def setup
+    @model_file = File.join(File.dirname(__FILE__), 'data', 'speaker1.gmm')
+  end
 
   def test_detection_threshold
     Diarize::Speaker.detection_threshold = 0.1
@@ -28,8 +31,7 @@ class SpeakerTest < Test::Unit::TestCase
   end
 
   def test_initialize_with_model
-    model_file = File.join(File.dirname(__FILE__), 'data', 'speaker1.gmm')
-    speaker = Diarize::Speaker.new(nil, nil, model_file)
+    speaker = Diarize::Speaker.new(nil, nil, @model_file)
     assert_equal speaker.model.name, 'S0'
   end
 
@@ -66,8 +68,7 @@ class SpeakerTest < Test::Unit::TestCase
   end
 
   def test_divergence_is_symmetric
-    model_file = File.join(File.dirname(__FILE__), 'data', 'speaker1.gmm')
-    speaker1 = Diarize::Speaker.new(nil, nil, model_file)
+    speaker1 = Diarize::Speaker.new(nil, nil, @model_file)
     speaker2 = Diarize::Speaker.ubm
     assert Diarize::Speaker.divergence(speaker1, speaker2) > 0
     assert_equal Diarize::Speaker.divergence(speaker1, speaker2), Diarize::Speaker.divergence(speaker2, speaker1)
@@ -75,16 +76,14 @@ class SpeakerTest < Test::Unit::TestCase
   end
 
   def test_divergence_ruby_is_same_as_divergence_lium
-    model_file = File.join(File.dirname(__FILE__), 'data', 'speaker1.gmm')
-    speaker1 = Diarize::Speaker.new(nil, nil, model_file)
+    speaker1 = Diarize::Speaker.new(nil, nil, @model_file)
     speaker2 = Diarize::Speaker.ubm
     assert_equal Diarize::Speaker.divergence_lium(speaker1, speaker2).round(12), Diarize::Speaker.divergence_ruby(speaker1, speaker2).round(12)
   end
 
-  def test_normalise
+  def test_normalize
     # Testing M-Norm
-    model_file = File.join(File.dirname(__FILE__), 'data', 'speaker1.gmm')
-    speaker1 = Diarize::Speaker.new(nil, nil, model_file)
+    speaker1 = Diarize::Speaker.new(nil, nil, @model_file)
     speaker2 = Diarize::Speaker.ubm
     assert Diarize::Speaker.divergence(speaker1, speaker2) != 1.0
     speaker1.normalize! # Putting speaker1.gmm at distance 1 from UBM
@@ -98,4 +97,38 @@ class SpeakerTest < Test::Unit::TestCase
     assert_equal old_supervector, speaker.supervector
   end
 
+  def test_to_rdf
+    speaker = Diarize::Speaker.new(URI.parse(""), "F", @model_file)
+    speaker.model_uri = URI.parse("https://www.example.com/model/1")
+    speaker.mean_log_likelihood = 0.9
+    to_rdf = speaker.to_rdf
+    assert_equal true, to_rdf.include?("ws:gender")
+    assert_equal true, to_rdf.include?("ws:model")
+    assert_equal true, to_rdf.include?("ws:mean_log_likelihood")
+    assert_equal true, to_rdf.include?("ws:supervector_hash")
+    assert_equal true, to_rdf.include?("https://www.example.com/model/1")
+  end
+
+  def test_as_json
+    speaker = Diarize::Speaker.new(URI.parse(""), "F", @model_file)
+    speaker.mean_log_likelihood = 0.9
+    as_json = speaker.as_json
+    assert_equal true, as_json.has_key?('gender')
+    assert_equal true, as_json.has_key?('model')
+    assert_equal true, as_json.has_key?('mean_log_likelihood')
+    assert_equal true, as_json.has_key?('supervector_hash')
+    assert_equal true, as_json.has_key?('supervector_sha')
+  end
+
+  def test_to_json
+    speaker    = Diarize::Speaker.new(URI.parse(""), "F", @model_file)
+    speaker.mean_log_likelihood = 0.9
+    to_json    = speaker.to_json
+    as_json    = JSON.parse(to_json)
+    assert_equal true, as_json.has_key?('gender')
+    assert_equal true, as_json.has_key?('model')
+    assert_equal true, as_json.has_key?('mean_log_likelihood')
+    assert_equal true, as_json.has_key?('supervector_hash')
+    assert_equal true, as_json.has_key?('supervector_sha')
+  end
 end

data/test/super_vector_test.rb

@@ -21,4 +21,16 @@ class SuperVectorTest < Test::Unit::TestCase
     assert_equal sv.vector.hash, sv.hash
   end
 
+  def test_sha
+    model = Diarize::Speaker.load_model(File.join(File.dirname(__FILE__), 'data', 'speaker1.gmm'))
+    sv = Diarize::SuperVector.generate_from_model(model)
+    assert_equal Digest::SHA256.hexdigest(sv.hash.to_s), sv.sha
+  end
+
+  def test_to_a
+    model = Diarize::Speaker.load_model(File.join(File.dirname(__FILE__), 'data', 'speaker1.gmm'))
+    sv = Diarize::SuperVector.generate_from_model(model)
+    assert_equal sv.instance_variable_get("@vector").to_a, sv.to_a
+  end
+
 end

data/test/version_test.rb

@@ -3,7 +3,7 @@ require 'test_helper'
 class VersionTest < Test::Unit::TestCase
 
   def test_current_version
-    assert_equal "0.3.7", Diarize::VERSION
+    assert_equal "0.3.8", Diarize::VERSION
   end
 
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: diarize-ruby
 version: !ruby/object:Gem::Version
-  version: 0.3.7
+  version: 0.3.8
 platform: ruby
 authors:
 - Yves Raimond
@@ -153,6 +153,7 @@ files:
 - lib/diarize/super_vector.rb
 - lib/diarize/ubm.gmm
 - lib/diarize/version.rb
+- lib/rjb/java_object_wrapper.rb
 - test/audio_test.rb
 - test/data/foo.wav
 - test/data/speaker1.gmm