word_aligner 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 9a4c26333b5e0991c70d0e61ef5435268672a112
+  data.tar.gz: 2ec6efcb6606676f2e30bd26425bd21843278018
+SHA512:
+  metadata.gz: 199f2c4f5288a60c907a9d842951e740ec712eeaca1a9371a74b96db92d2053107543809b107c23bd025f532f5a75add474f5ff7f12cd163d0c99cb728cb5ffd
+  data.tar.gz: 95ae32a917c4e2578bb6087973c0cf195eb4b41b56e771b37d1d138b336440163e8bee9e1ebf015bb48b9690831654ed08fcce8c962dcbae3f11d765f9f01d86

data/Gemfile

@@ -0,0 +1,18 @@
+source "http://rubygems.org"
+# Add dependencies required to use your gem here.
+# Example:
+#   gem "activesupport", ">= 2.3.5"
+
+# Add dependencies to develop your gem here.
+# Include everything needed to run rake, tests, features, etc.
+group :development do
+  gem "guard"
+  gem "guard-rspec"
+  gem "shoulda", ">= 0"
+  gem "rdoc", "~> 3.12"
+  gem "bundler", "~> 1.0"
+  gem "jeweler", "~> 1.8.7"
+  gem "unicode_utils"
+  gem "rspec-core"
+  gem "guard-rspec"
+end

data/Gemfile.lock

@@ -0,0 +1,115 @@
+GEM
+  remote: http://rubygems.org/
+  specs:
+    activesupport (4.0.0)
+      i18n (~> 0.6, >= 0.6.4)
+      minitest (~> 4.2)
+      multi_json (~> 1.3)
+      thread_safe (~> 0.1)
+      tzinfo (~> 0.3.37)
+    addressable (2.3.5)
+    atomic (1.1.13)
+    builder (3.2.2)
+    coderay (1.0.9)
+    diff-lcs (1.2.4)
+    faraday (0.8.8)
+      multipart-post (~> 1.2.0)
+    ffi (1.9.0)
+    formatador (0.2.4)
+    git (1.2.6)
+    github_api (0.10.1)
+      addressable
+      faraday (~> 0.8.1)
+      hashie (>= 1.2)
+      multi_json (~> 1.4)
+      nokogiri (~> 1.5.2)
+      oauth2
+    guard (1.8.2)
+      formatador (>= 0.2.4)
+      listen (>= 1.0.0)
+      lumberjack (>= 1.0.2)
+      pry (>= 0.9.10)
+      thor (>= 0.14.6)
+    guard-rspec (3.0.3)
+      guard (>= 1.8)
+      rspec (~> 2.13)
+    hashie (2.0.5)
+    highline (1.6.19)
+    httpauth (0.2.0)
+    i18n (0.6.5)
+    jeweler (1.8.7)
+      builder
+      bundler (~> 1.0)
+      git (>= 1.2.5)
+      github_api (= 0.10.1)
+      highline (>= 1.6.15)
+      nokogiri (= 1.5.10)
+      rake
+      rdoc
+    json (1.8.0)
+    jwt (0.1.8)
+      multi_json (>= 1.5)
+    listen (1.3.1)
+      rb-fsevent (>= 0.9.3)
+      rb-inotify (>= 0.9)
+      rb-kqueue (>= 0.2)
+    lumberjack (1.0.4)
+    method_source (0.8.2)
+    minitest (4.7.5)
+    multi_json (1.8.0)
+    multi_xml (0.5.5)
+    multipart-post (1.2.0)
+    nokogiri (1.5.10)
+    oauth2 (0.9.2)
+      faraday (~> 0.8)
+      httpauth (~> 0.2)
+      jwt (~> 0.1.4)
+      multi_json (~> 1.0)
+      multi_xml (~> 0.5)
+      rack (~> 1.2)
+    pry (0.9.12.2)
+      coderay (~> 1.0.5)
+      method_source (~> 0.8)
+      slop (~> 3.4)
+    rack (1.5.2)
+    rake (10.1.0)
+    rb-fsevent (0.9.3)
+    rb-inotify (0.9.1)
+      ffi (>= 0.5.0)
+    rb-kqueue (0.2.0)
+      ffi (>= 0.5.0)
+    rdoc (3.12.2)
+      json (~> 1.4)
+    rspec (2.14.1)
+      rspec-core (~> 2.14.0)
+      rspec-expectations (~> 2.14.0)
+      rspec-mocks (~> 2.14.0)
+    rspec-core (2.14.5)
+    rspec-expectations (2.14.2)
+      diff-lcs (>= 1.1.3, < 2.0)
+    rspec-mocks (2.14.3)
+    shoulda (3.5.0)
+      shoulda-context (~> 1.0, >= 1.0.1)
+      shoulda-matchers (>= 1.4.1, < 3.0)
+    shoulda-context (1.1.5)
+    shoulda-matchers (2.3.0)
+      activesupport (>= 3.0.0)
+    slop (3.4.6)
+    thor (0.18.1)
+    thread_safe (0.1.2)
+      atomic
+    tzinfo (0.3.37)
+    unicode_utils (1.4.0)
+
+PLATFORMS
+  ruby
+
+DEPENDENCIES
+  bundler (~> 1.0)
+  guard
+  guard-rspec
+  jeweler (~> 1.8.7)
+  rdoc (~> 3.12)
+  rspec-core
+  shoulda
+  unicode_utils

data/Guardfile

@@ -0,0 +1,8 @@
+# A sample Guardfile
+# More info at https://github.com/guard/guard#readme
+
+guard :rspec, cli: '--color --format nested' do
+  watch(%r{^spec/.+_spec\.rb$})
+  watch(%r{^lib/(.+)\.rb$})     { |m| "spec/lib/#{m[1]}_spec.rb" }
+  watch('spec/spec_helper.rb')  { "spec" }
+end

data/LICENSE.txt

@@ -0,0 +1,20 @@
+Copyright (c) 2013 Lunatyq, skpvox
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

data/README.md

@@ -0,0 +1,45 @@
+# word_aligner
+
+word_aligner lets you compare string as word_align.pl script does in cmu-sphinx.
+
+It calculates distance in terms of insertions, deletions and substitutions.
+
+## Installation
+
+Add this line to your Gemfile:
+
+```
+gem 'word_aligner'
+```
+
+Then execute:
+
+```
+$ bundle
+```
+
+## Usage
+
+```
+ error_rate = WordAligner.align('example shown line', 'they shown line')
+ error_rate.aligned_transcription => "EXAMPLE shown line"
+ error_rate.aligned_hypothesis => "THEY    shown line"
+ error_rate.words => 3
+ error_rate.insertions => 0
+ error_rate.deletions => 0
+ error_rate.substitutions => 1
+
+ error_rate.correct => 2
+ error_rate.errors => 1
+ error_rate.percent_correct  => 66.0
+ error_rate.percent_error    => 33.0
+ error_rate.percent_accuracy => 66.0
+```
+
+## Contributing
+
+1. Fork it
+2. Create your feature branch (`git checkout -b my-new-feature`)
+3. Commit your changes (`git commit -am 'Add some feature'`)
+4. Push to the branch (`git push origin my-new-feature`)
+5. Create new Pull Request

data/Rakefile

@@ -0,0 +1,45 @@
+# encoding: utf-8
+
+require 'rubygems'
+require 'bundler'
+begin
+  Bundler.setup(:default, :development)
+rescue Bundler::BundlerError => e
+  $stderr.puts e.message
+  $stderr.puts "Run `bundle install` to install missing gems"
+  exit e.status_code
+end
+require 'rake'
+
+require 'jeweler'
+Jeweler::Tasks.new do |gem|
+  # gem is a Gem::Specification... see http://docs.rubygems.org/read/chapter/20 for more options
+  gem.name = "word_aligner"
+  gem.homepage = "http://github.com/lunatyq/word_aligner"
+  gem.license = "MIT"
+  gem.summary = %Q{TODO: one-line summary of your gem}
+  gem.description = %Q{TODO: longer description of your gem}
+  gem.email = "maciej@szukio.pl"
+  gem.authors = ["Lunatyq"]
+  # dependencies defined in Gemfile
+end
+Jeweler::RubygemsDotOrgTasks.new
+
+require 'rake/testtask'
+Rake::TestTask.new(:test) do |test|
+  test.libs << 'lib' << 'test'
+  test.pattern = 'test/**/test_*.rb'
+  test.verbose = true
+end
+
+task :default => :test
+
+require 'rdoc/task'
+Rake::RDocTask.new do |rdoc|
+  version = File.exist?('VERSION') ? File.read('VERSION') : ""
+
+  rdoc.rdoc_dir = 'rdoc'
+  rdoc.title = "word_aligner #{version}"
+  rdoc.rdoc_files.include('README*')
+  rdoc.rdoc_files.include('lib/**/*.rb')
+end

data/VERSION

@@ -0,0 +1 @@
+0.1.0

data/lib/word_aligner.rb

@@ -0,0 +1,11 @@
+require 'word_aligner/aligner'
+require 'word_aligner/word_error_rate'
+
+module WordAligner
+
+  module_function
+
+  def align(transcription, hypothesis)
+    Aligner.new(transcription, hypothesis).word_error_rate
+  end
+end

data/lib/word_aligner/aligner.rb

@@ -0,0 +1,170 @@
+require 'unicode_utils/upcase'
+
+module WordAligner
+  class Aligner
+
+    INSERTION  = 1
+    DELETION   = 2
+    MATCHING   = 3
+    SUBSTITUTE = 4
+
+  	attr_accessor :insertions, :substitutions, :deletions, :matching, :align_cost
+  	attr_accessor :aligned_transcription, :aligned_hypothesis
+
+    attr_reader :transcription,    :hypothesis
+    attr_reader :reference_tokens, :hypothese_tokens
+
+    attr_reader :align_matrix, :backtrace_matrix
+
+    def initialize(transcription, hypothesis)
+      @transcription = transcription
+      @hypothesis    = hypothesis
+
+  		@insertions    = 0
+  		@substitutions = 0
+  		@deletions     = 0
+  		@matching      = 0
+
+      @reference_tokens = tokenize(transcription)
+      @hypothese_tokens = tokenize(hypothesis)
+    end
+
+    def word_error_rate
+      align_sentences if align_matrix.nil?
+
+      WordErrorRate.new(result)
+    end
+
+    def self.align(original, hypothesis)
+      new(original, hypothesis).result
+    end
+
+  	private
+
+    def result
+  		{
+        aligned_transcription: aligned_transcription,
+        aligned_hypothesis:    aligned_hypothesis,
+        transcription:         transcription,
+        hypothesis:            hypothesis,
+        transcription_words:   reference_tokens.size,
+        insertions:            insertions,
+        deletions:             deletions,
+        substitutions:         substitutions,
+        matching:              matching,
+        align_cost:            align_cost
+      }
+    end
+
+    def align_sentences
+      initialize_matrices
+
+      align_tokens
+
+  		generate_aligned_sentences
+    end
+
+  	def generate_aligned_sentences
+      backtrace.each do |ref, hyp|
+        width = 0
+
+        if ref && hyp and (ref == hyp)
+          ref = ref.downcase
+          hyp = hyp.downcase
+        end
+
+        ref ||= '***'
+        hyp ||= '***'
+
+        width = [ref.size, hyp.size, width, 3].max
+
+        aligned_transcription << '%-*s ' % [width, ref]
+        aligned_hypothesis    << '%-*s ' % [width, hyp]
+      end
+
+  		aligned_transcription.strip!
+  		aligned_hypothesis.strip!
+  	end
+
+    def initialize_matrices
+      @align_matrix     = []
+      @backtrace_matrix = []
+
+      align_matrix     << (0..hypothese_tokens.size).to_a
+
+      backtrace_matrix << Array.new(hypothese_tokens.size+1, INSERTION)
+
+      (0..reference_tokens.size).each { |j| align_matrix[j]     ||= []; align_matrix[j][0]     = j }
+      (0..reference_tokens.size).each { |i| backtrace_matrix[i] ||= []; backtrace_matrix[i][0] = DELETION }
+
+  		self.aligned_transcription = ''
+      self.aligned_hypothesis    = ''
+    end
+
+    def align_tokens
+      (1..reference_tokens.size).each do |i|
+        (1..hypothese_tokens.size).each do |j|
+          cost  = reference_tokens[i - 1] != hypothese_tokens[j - 1] ? 1 : 0
+
+          ins   = align_matrix[i][j - 1] + 1
+          del   = align_matrix[i - 1][j] + 1
+          subst = align_matrix[i - 1][j - 1] + cost
+          min   = [ins,del,subst].min
+
+          align_matrix[i][j] = min
+
+          if min == subst
+            backtrace_matrix[i][j] = MATCHING + cost
+          elsif min == ins
+            backtrace_matrix[i][j] = INSERTION
+          elsif min == del
+            backtrace_matrix[i][j] = DELETION
+          end
+        end
+      end
+
+      self.align_cost = align_matrix[reference_tokens.size][hypothese_tokens.size]
+    end
+
+    def backtrace
+      i = reference_tokens.size
+      j = hypothese_tokens.size
+
+      alignment = []
+
+      while !(i == 0 and j == 0)
+        pointer = backtrace_matrix[i][j]
+
+        case pointer
+        when INSERTION
+  	      alignment.unshift [nil, hypothese_tokens[j - 1]]
+  				self.insertions += 1
+  				j -= 1
+        when DELETION
+          alignment.unshift [reference_tokens[i - 1], nil]
+          self.deletions += 1
+          i -= 1
+        when MATCHING
+          alignment.unshift [reference_tokens[i - 1], hypothese_tokens[j - 1]]
+          self.matching += 1
+          j -= 1
+          i -= 1
+        when SUBSTITUTE
+          alignment.unshift [reference_tokens[i - 1], hypothese_tokens[j - 1]]
+          self.substitutions += 1
+          j -= 1
+          i -= 1
+        else
+          break
+        end
+      end
+
+      alignment
+    end
+
+    def tokenize(word)
+      UnicodeUtils.upcase(word.rstrip).split(' ')
+    end
+
+  end
+end