phonetics 1.5.4 → 1.8.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e8f26a3fc9c4a6729a0ddd04beee049efc4d1c0e8e2b887cfd14974de71d989a
-  data.tar.gz: 4449ecff87444214065a0f964c2cd7a0ba2ce93f52b3cbeae14e4777afcba3df
+  metadata.gz: '05877cd48ad83d92a29dbcf60a5bd73373c7ecf974387487d37b719e4df0db4d'
+  data.tar.gz: d400d6249ac9204e920e8bc617b00b21009a4d15aeb719e589a100482114a927
 SHA512:
-  metadata.gz: 4309942350149685324fa56a9c642f9456c7abe1993ab7f0801fbf56883afc4e31750e969e6d49e3da0ecd7b16a8d93b9212f5469ddc1dcd4c86699d0f48cd46
-  data.tar.gz: 8007ae8ef77e2d9e5f055fc597c4c5ffbd9ee00007d9d31ed4d06ae96e9980bc2dfc92149c96bd94098faa7d0670c5d7cbd13a30f45acd8d8c6000853692f341
+  metadata.gz: 7bfd7f82c7f579e377f8d7669480c7cf4d73cf1c1b1259db1e19d4bbb9c8573480a7825ca51df6e4f600feba67ecfee10a2f9b66a49a9f33e888f94b04be9528
+  data.tar.gz: 3ddc3f72fbec2a0f833512b1d84c6d05d30c90db63e48ceb7f44c19cd7df43205517079b4cda113e9d80fe1a1c27586771eaf575c423beec2fa46a55a504cfd0

data/Rakefile

@@ -5,23 +5,34 @@ require 'rake/extensiontask'
 require 'rspec/core/rake_task'
 require 'rubocop/rake_task'
 
+EXT_PATH = 'ext/c_levenshtein'
+
 Rake::ExtensionTask.new('c_levenshtein') do |extension|
-  extension.ext_dir = 'ext/c_levenshtein'
+  extension.ext_dir = EXT_PATH
   extension.lib_dir = 'lib/phonetics'
 end
 
-PHONETIC_COST_C_EXTENSION = File.expand_path('ext/c_levenshtein/phonetic_cost.c', __dir__)
+PHONETIC_COST_C_EXTENSION = File.join(EXT_PATH, 'phonetic_cost.c')
+NEXT_PHONEME_LENGTH_C_EXTENSION = File.join(EXT_PATH, 'next_phoneme_length.c')
+
+require_relative './lib/phonetics/code_generator'
 
-namespace :compile do
-  desc 'Write phonetic_cost.c using Phonetic values'
-  task :phonetic_cost do
-    require_relative './lib/phonetics'
-    file = File.open(PHONETIC_COST_C_EXTENSION, 'w')
-    Phonetics.generate_phonetic_cost_c_code(file)
-    puts "Wrote #{PHONETIC_COST_C_EXTENSION}"
-  end
+desc 'Write phonetic_cost.c using Phonetic values'
+task PHONETIC_COST_C_EXTENSION do
+  file = File.open(PHONETIC_COST_C_EXTENSION, 'w')
+  Phonetics::CodeGenerator.new(file).generate_phonetic_cost_c_code
+  puts "Wrote #{PHONETIC_COST_C_EXTENSION}"
 end
-task compile: 'compile:phonetic_cost'
+
+desc 'Write phonemes.c using a lookup table of byte arrays'
+task NEXT_PHONEME_LENGTH_C_EXTENSION do
+  file = File.open(NEXT_PHONEME_LENGTH_C_EXTENSION, 'w')
+  Phonetics::CodeGenerator.new(file).generate_next_phoneme_length_c_code
+  puts "Wrote #{NEXT_PHONEME_LENGTH_C_EXTENSION}"
+end
+
+task compile: PHONETIC_COST_C_EXTENSION
+task compile: NEXT_PHONEME_LENGTH_C_EXTENSION
 
 RSpec::Core::RakeTask.new(:spec)
 

data/VERSION

@@ -1 +1 @@
-1.5.4
+1.8.0

data/bin/console

@@ -0,0 +1,13 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+$LOAD_PATH << '../lib'
+
+require 'bundler/setup'
+require 'pry-byebug'
+require 'phonetics'
+require 'phonetics/levenshtein'
+require 'phonetics/ruby_levenshtein'
+require 'phonetics/code_generator'
+
+Phonetics.pry

data/ext/c_levenshtein/levenshtein.c

@@ -1,5 +1,9 @@
-#include "ruby.h"
 #include <stdbool.h>
+#include "ruby.h"
+#include "ruby/encoding.h"
+#include "ruby/re.h"
+#include "./phonemes.h"
+#include "./next_phoneme_length.h"
 #include "./phonetic_cost.h"
 
 #define debug(M, ...) if (verbose) printf(M, ##__VA_ARGS__)
@@ -10,8 +14,8 @@ VALUE Binding = Qnil;
 
 void Init_c_levenshtein();
 
-void set_initial(double *d, int *string1, int string1_length, int *string2, int string2_length, bool verbose);
-void print_matrix(double *d, int *string1, int string1_length, int *string2, int string2_length, bool verbose);
+void set_initial(double *d, int *string1, int string1_phoneme_count, int *string1_phoneme_sizes, int *string2, int string2_phoneme_count, int *string2_phoneme_sizes, bool verbose);
+void print_matrix(double *d, int *string1, int string1_phoneme_count, int *string1_phoneme_sizes, int *string2, int string2_phoneme_count, int *string2_phoneme_sizes, bool verbose);
 VALUE method_internal_phonetic_distance(VALUE self, VALUE _string1, VALUE _string2, VALUE _verbose);
 
 /* Function implemitations */
@@ -22,15 +26,23 @@ void Init_c_levenshtein() {
 }
 
 VALUE method_internal_phonetic_distance(VALUE self, VALUE _string1, VALUE _string2, VALUE _verbose){
+  if (!RB_TYPE_P(_string1, T_STRING)) {
+    rb_raise(rb_eArgError, "must pass string as first argument");
+  }
+  if (!RB_TYPE_P(_string2, T_STRING)) {
+    rb_raise(rb_eArgError, "must pass string as second argument");
+  }
 
-  VALUE *string1_ruby = RARRAY_PTR(_string1);
-  VALUE *string2_ruby = RARRAY_PTR(_string2);
   bool verbose = _verbose;
-  int string1_length = (int) RARRAY_LEN(_string1);
-  int string2_length = (int) RARRAY_LEN(_string2);
-  // We name them as 'strings' but in C-land we're representing our strings as
-  // arrays of `int`s, where each int represents a consistent (if unusual)
-  // encoding of a grapheme cluster (a symbol for a phoneme).
+  int string1_length = (int) RSTRING_LEN(_string1);
+  int string2_length = (int) RSTRING_LEN(_string2);
+
+  // Given the input strings, we count the phonemes in each and store both the
+  // total and, in a phoneme_sizes array, the length of each.
+  int string1_phoneme_count = 0;
+  int string2_phoneme_count = 0;
+  int string1_phoneme_sizes[string1_length + 1];
+  int string2_phoneme_sizes[string2_length + 1];
   int string1[string1_length + 1];
   int string2[string2_length + 1];
 
@@ -42,86 +54,85 @@ VALUE method_internal_phonetic_distance(VALUE self, VALUE _string1, VALUE _strin
          insert, replace,
          cost;
   int i, j;               // Frequently overwritten loop vars
-
-  // Guard clause for two empty strings
-  if (string1_length == 0 && string2_length == 0)
-    return DBL2NUM(0.0);
-
-  //
-  // Intial data setup
-  //
+  int string1_offset = 0;
+  int string2_offset = 0;
 
   for (i = 0; i < string1_length; i++) {
-    string1[i] = NUM2INT(string1_ruby[i]);
-    debug("string1[%d] = %d\n", i, string1[i]);
+    string1[i] = (RSTRING_PTR(_string1)[i] & 0xff);
   }
-  for (j = 0; j < string2_length; j++) {
-    string2[j] = NUM2INT(string2_ruby[j]);
-    debug("string2[%d] = %d\n", i, string2[j]);
+  for (i = 0; i < string2_length; i++) {
+    string2[i] = RSTRING_PTR(_string2)[i] & 0xff;
   }
 
-  // one-dimensional representation of 2 dimentional array len(string1)+1 *
-  // len(string2)+1
-  d = malloc((sizeof(double)) * (string1_length+1) * (string2_length+1));
+  find_phonemes(string1, string1_length, &string1_phoneme_count, string1_phoneme_sizes);
+  find_phonemes(string2, string2_length, &string2_phoneme_count, string2_phoneme_sizes);
 
-  //
-  // Fill in the (flattened) matrix using the Levenshtein algorithm so we can
-  // pluck the lowest-cost edit distance (stored in the lower-right corner, in
-  // this case the last spot in the array)
-  //
+  // Guard clauses for empty strings
+  if (string1_phoneme_count == 0 && string2_phoneme_count == 0)
+    return DBL2NUM(0.0);
   
+  // one-dimensional representation of 2 dimensional array
+  d = malloc((sizeof(double)) * (string1_phoneme_count+1) * (string2_phoneme_count+1));
+
   // First, set the top row and left column of the matrix using the sequential
   // phonetic edit distance of string1 and string2, respectively
-  set_initial(d, string1, string1_length, string2, string2_length, verbose);
+  set_initial(d, string1, string1_phoneme_count, string1_phoneme_sizes, string2, string2_phoneme_count, string2_phoneme_sizes, verbose);
 
-  debug("before:\n");
-  print_matrix(d, string1, string1_length, string2, string2_length, verbose);
+  print_matrix(d, string1, string1_phoneme_count, string1_phoneme_sizes, string2, string2_phoneme_count, string2_phoneme_sizes, verbose);
 
-  // Then walk through the matrix and fill in each cell with the lowest-cost
-  // phonetic edit distance for that matrix cell.
+  // Then Fill in the (flattened) matrix using the Levenshtein algorithm so we can
+  // pluck the lowest-cost edit distance (stored in the lower-right corner, in
+  // this case the last spot in the array).
+  // We'll use phonetic distance instead of '1' as the edit cost.
+  //
   // (Skipping i=0 and j=0 because set_initial filled in all cells where i
   // or j are zero-valued)
-  for (j = 1; j <= string2_length; j++){
-    for (i = 1; i <= string1_length; i++){
+  for (j = 1; j <= string2_phoneme_count; j++){
+
+    for (i = 1; i <= string1_phoneme_count; i++){
 
       // The cost of deletion or addition is the Levenshtein distance
       // calculation (the value in the cell to the left, upper-left, or above)
       // plus the phonetic distance between the sound we're moving from to the
       // new one.
 
-      debug("------- %d/%d (%d) \n", i, j, j*(string1_length+1) + i);
+      debug("------- %d/%d (%d) \n", i, j, j*(string1_phoneme_count+1) + i);
 
-      cost = phonetic_cost(string1[i-1], string2[j-1]);
-      debug("phonetic cost of %d to %d is %f\n", string1[i-1], string2[j-1], cost);
+      // TODO: increment i and j by the phoneme lengths
+      cost = phonetic_cost(string1, string1_offset, string1_phoneme_sizes[i-1], string2, string2_offset, string2_phoneme_sizes[j-1]);
 
-      insert = d[j*(string1_length+1) + i-1];
+      insert = d[j*(string1_phoneme_count+1) + i-1];
       debug("insert proposes cell %d,%d - %f\n", i-1, j, insert);
       min = insert;
       debug("min (insert): %f\n", min);
 
-      delete = d[(j-1)*(string1_length+1) + i];
+      delete = d[(j-1)*(string1_phoneme_count+1) + i];
       debug("delete proposes cell %d,%d - %f\n", i, j-1, delete);
       if (delete < min) {
         debug("delete is %f, better than %f for %d/%d\n", delete, min, i, j);
         min = delete;
       }
 
-      replace = d[(j-1)*(string1_length+1) + i-1];
+      replace = d[(j-1)*(string1_phoneme_count+1) + i-1];
       debug("replace proposes cell %d,%d - %f\n", i-1, j-1, replace);
       if (replace < min) {
         debug("replace is %f, better than %f for %d/%d\n", replace, min, i, j);
         min = replace;
       }
 
-      d[(j * (string1_length+1)) + i] = min + cost;
+      d[(j * (string1_phoneme_count+1)) + i] = min + cost;
       debug("\n");
-      print_matrix(d, string1, string1_length, string2, string2_length, verbose);
+      print_matrix(d, string1, string1_phoneme_count, string1_phoneme_sizes, string2, string2_phoneme_count, string2_phoneme_sizes, verbose);
+
+      string1_offset += string1_phoneme_sizes[i-1];
     }
+    string1_offset = 0;
+    string2_offset += string2_phoneme_sizes[j-1];
   }
 
   // The final element in the `d` array is the value of the shortest path from
   // the top-left to the bottom-right of the matrix.
-  distance = d[(string1_length + 1) * (string2_length + 1) - 1];
+  distance = d[(string1_phoneme_count + 1) * (string2_phoneme_count + 1) - 1];
 
   free(d);
   debug("distance: %f\n", distance);
@@ -129,7 +140,6 @@ VALUE method_internal_phonetic_distance(VALUE self, VALUE _string1, VALUE _strin
   return DBL2NUM(distance);
 }
 
-
 // Set the minimum scores equal to the distance between each phoneme,
 // sequentially.
 //
@@ -139,12 +149,14 @@ VALUE method_internal_phonetic_distance(VALUE self, VALUE _string1, VALUE _strin
 // Subsequent values are the cumulative phonetic distance between each
 // phoneme within the same string.
 // "aek" -> [0.0, 1.0, 1.61, 2.61]
-void set_initial(double *d, int *string1, int string1_length, int *string2, int string2_length, bool verbose) {
+void set_initial(double *d, int *string1, int string1_phoneme_count, int *string1_phoneme_sizes, int *string2, int string2_phoneme_count, int *string2_phoneme_sizes, bool verbose) {
 
   double initial_distance;
+  int string1_offset = 0;
+  int string2_offset = 0;
   int i, j;
 
-  if (string1_length == 0 || string2_length == 0) {
+  if (string1_phoneme_count == 0 || string2_phoneme_count == 0) {
     initial_distance = 0.0;
   } else {
     initial_distance = 1.0;
@@ -152,51 +164,67 @@ void set_initial(double *d, int *string1, int string1_length, int *string2, int
 
   // The top-left is 0, the cell to the right and down are each 1 to start
   d[0] = (double) 0.0;
-  if (string1_length > 0) {
+  if (string1_phoneme_count > 0) {
     d[1] = initial_distance;
   }
-  if (string2_length > 0) {
-    d[string1_length+1] = initial_distance;
+  if (string2_phoneme_count > 0) {
+    d[string1_phoneme_count+1] = initial_distance;
   }
 
-  debug("string1 length: %d\n", string1_length);
-  for (i=2; i <= string1_length; i++) {
+  debug("string1 phoneme count: %d\n", string1_phoneme_count);
+
+  for (i=2; i <= string1_phoneme_count; i++) {
     // The cost of adding the next phoneme is the cost so far plus the phonetic
     // distance between the previous one and the current one.
-    d[i] = d[i-1] + phonetic_cost(string1[i-2], string1[i-1]);
+    d[i] = d[i-1] +
+      phonetic_cost(string1, string1_offset, string1_phoneme_sizes[i-2], string1, string1_offset + string1_phoneme_sizes[i-2], string1_phoneme_sizes[i-1]);
+    string1_offset += string1_phoneme_sizes[i-2];
   }
-  debug("string2 length: %d\n", string2_length);
-  for (j=2; j <= string2_length; j++) {
+
+  debug("string2 phoneme count: %d\n", string2_phoneme_count);
+
+  for (j=2; j <= string2_phoneme_count; j++) {
     // The same exact pattern down the left side of the matrix
-    d[j * (string1_length+1)] = d[(j - 1) * (string1_length+1)] + phonetic_cost(string2[j-2], string2[j-1]);
+    d[j * (string1_phoneme_count+1)] = d[(j - 1) * (string1_phoneme_count+1)] +
+      phonetic_cost(string2, string2_offset, string2_phoneme_sizes[j-2], string2, string2_offset + string2_phoneme_sizes[j-2], string2_phoneme_sizes[j-1]);
+    string2_offset += string2_phoneme_sizes[j-1];
   }
 
-  // And zero out the rest. If you're reading this please edit this to be
-  // faster.
-  for (j=1; j <= string2_length; j++) {
-    for (i=1; i <= string1_length; i++) {
-      d[j * (string1_length+1) + i] = (double) 0.0;
+  // And zero out the rest. If you're reading this please show me a way to do
+  // this faster.
+  for (j=1; j <= string2_phoneme_count; j++) {
+    for (i=1; i <= string1_phoneme_count; i++) {
+      d[j * (string1_phoneme_count+1) + i] = (double) 0.0;
     }
   }
 }
 
 // A handy visualization for developers
-void print_matrix(double *d, int *string1, int string1_length, int *string2, int string2_length, bool verbose) {
+void print_matrix(double *d, int *string1, int string1_phoneme_count, int *string1_phoneme_sizes, int *string2, int string2_phoneme_count, int *string2_phoneme_sizes, bool verbose) {
+
   int i, j;
-  debug("              ");
-  for (i=0; i < string1_length; i++) {
-    debug("%8.d ", string1[i]);
+  int string1_offset = 0;
+  int string2_offset = 0;
+
+  if (!verbose)
+    return;
+
+  printf("           ");
+  for (i=0; i < string1_phoneme_count; i++) {
+    print_phoneme(string1, string1_offset, string1_phoneme_sizes[i], 9);
+    string1_offset += string1_phoneme_sizes[i];
   }
-  debug("\n");
-  for (j=0; j <= string2_length; j++) {
+  printf("\n");
+  for (j=0; j <= string2_phoneme_count; j++) {
     if (j==0) {
-      debug("     ");
+      printf("  ");
     } else {
-      debug("%4.d ", string2[j-1]);
-    }
-    for (i=0; i <= string1_length; i++) {
-      debug("%f ", d[j * (string1_length+1) + i]) ;
+      print_phoneme(string2, string2_offset, string2_phoneme_sizes[j-1], 2);
+      string2_offset += string2_phoneme_sizes[j-1];
+    } 
+    for (i=0; i <= string1_phoneme_count; i++) {
+      printf("%f ", d[j * (string1_phoneme_count+1) + i]) ;
     }
-    debug("\n");
+    printf("\n");
   }
 }

data/ext/c_levenshtein/next_phoneme_length.c

@@ -0,0 +1,1364 @@
+// This is compiled from Ruby, in phonetics/lib/phonetics/code_generator.rb:221
+int next_phoneme_length(int *string, int cursor, int length) {
+
+  int max_length;
+  max_length = length - cursor;
+
+    switch(string[cursor + 0]) {
+
+    case 105:
+      // Phoneme: "i", bytes: [105]
+      // vowel features: {"F1":240,"F2":2400,"rounded":false}
+     return 1;
+        break;
+    case 121:
+      // Phoneme: "y", bytes: [121]
+      // vowel features: {"F1":235,"F2":2100,"rounded":false}
+     return 1;
+        break;
+    case 201:
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 170:
+        // Phoneme: "ɪ", bytes: [201, 170]
+        // vowel features: {"F1":300,"F2":2100,"rounded":false}
+       return 2;
+          break;
+      case 155:
+        // Phoneme: "ɛ", bytes: [201, 155]
+        // vowel features: {"F1":610,"F2":1900,"rounded":false}
+       return 2;
+          break;
+      case 182:
+        // Phoneme: "ɶ", bytes: [201, 182]
+        // vowel features: {"F1":820,"F2":1530,"rounded":true}
+       return 2;
+          break;
+      case 145:
+        // Phoneme: "ɑ", bytes: [201, 145]
+        // vowel features: {"F1":750,"F2":940,"rounded":false}
+       return 2;
+          break;
+      case 146:
+        // Phoneme: "ɒ", bytes: [201, 146]
+        // vowel features: {"F1":700,"F2":760,"rounded":true}
+       return 2;
+          break;
+      case 153:
+        // Phoneme: "ə", bytes: [201, 153]
+        // vowel features: {"F1":600,"F2":1170,"rounded":false}
+       return 2;
+          break;
+      case 157:
+        // Phoneme: "ɝ", bytes: [201, 157]
+        // vowel features: {"F1":600,"F2":1170,"rounded":false,"rhotic":true}
+       return 2;
+          break;
+      case 148:
+        // Phoneme: "ɔ", bytes: [201, 148]
+        // vowel features: {"F1":500,"F2":700,"rounded":true}
+       return 2;
+          break;
+      case 164:
+        // Phoneme: "ɤ", bytes: [201, 164]
+        // vowel features: {"F1":460,"F2":1310,"rounded":false}
+       return 2;
+          break;
+      case 175:
+        // Phoneme: "ɯ", bytes: [201, 175]
+        // vowel features: {"F1":300,"F2":1390,"rounded":false}
+       return 2;
+          break;
+      case 177:
+        // Phoneme: "ɱ", bytes: [201, 177]
+        // consonant features: {"position":"Labio-dental","position_index":2,"manner":"Nasal","voiced":true}
+       return 2;
+          break;
+      case 179:
+        // Phoneme: "ɳ", bytes: [201, 179]
+        // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Nasal","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 138:
+            // Phoneme: "ɳ̊", bytes: [201, 179, 204, 138]
+            // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Nasal","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 178:
+        // Phoneme: "ɲ", bytes: [201, 178]
+        // consonant features: {"position":"Palatal","position_index":8,"manner":"Nasal","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 138:
+            // Phoneme: "ɲ̊", bytes: [201, 178, 204, 138]
+            // consonant features: {"position":"Palatal","position_index":8,"manner":"Nasal","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 180:
+        // Phoneme: "ɴ", bytes: [201, 180]
+        // consonant features: {"position":"Uvular","position_index":10,"manner":"Nasal","voiced":true}
+       return 2;
+          break;
+      case 150:
+        // Phoneme: "ɖ", bytes: [201, 150]
+        // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Stop","voiced":true}
+       return 2;
+          break;
+      case 159:
+        // Phoneme: "ɟ", bytes: [201, 159]
+        // consonant features: {"position":"Palatal","position_index":8,"manner":"Stop","voiced":true}
+       return 2;
+          break;
+      case 162:
+        // Phoneme: "ɢ", bytes: [201, 162]
+        // consonant features: {"position":"Uvular","position_index":10,"manner":"Stop","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 134:
+            // Phoneme: "ɢ̆", bytes: [201, 162, 204, 134]
+            // consonant features: {"position":"Uvular","position_index":10,"manner":"Tap/flap","voiced":true}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 149:
+        // Phoneme: "ɕ", bytes: [201, 149]
+        // consonant features: {"position":"Palatal","position_index":8,"manner":"Sibilant fricative","voiced":false}
+       return 2;
+          break;
+      case 184:
+        // Phoneme: "ɸ", bytes: [201, 184]
+        // consonant features: {"position":"Bi-labial","position_index":1,"manner":"Non-sibilant fricative","voiced":false}
+       return 2;
+          break;
+      case 185:
+        // Phoneme: "ɹ", bytes: [201, 185]
+        // consonant features: {"position":"Alveolar","position_index":5,"manner":"Approximant","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 160:
+           if (max_length > 4) {
+            switch(string[cursor + 4]) {
+
+            case 204:
+             if (max_length > 5) {
+              switch(string[cursor + 5]) {
+
+              case 138:
+               if (max_length > 6) {
+                switch(string[cursor + 6]) {
+
+                case 203:
+                 if (max_length > 7) {
+                  switch(string[cursor + 7]) {
+
+                  case 148:
+                    // Phoneme: "ɹ̠̊˔", bytes: [201, 185, 204, 160, 204, 138, 203, 148]
+                    // consonant features: {"position":"Post-alveolar","position_index":6,"manner":"Non-sibilant fricative","voiced":false}
+                   return 8;
+                      break;
+                  }
+                 } else {
+                   return 7;
+                 }
+                    break;
+                }
+               } else {
+                 return 6;
+               }
+                  break;
+              }
+             } else {
+               return 5;
+             }
+                break;
+            case 203:
+             if (max_length > 5) {
+              switch(string[cursor + 5]) {
+
+              case 148:
+                // Phoneme: "ɹ̠˔", bytes: [201, 185, 204, 160, 203, 148]
+                // consonant features: {"position":"Post-alveolar","position_index":6,"manner":"Non-sibilant fricative","voiced":true}
+               return 6;
+                  break;
+              }
+             } else {
+               return 5;
+             }
+                break;
+            }
+           } else {
+             return 4;
+           }
+              break;
+          case 165:
+            // Phoneme: "ɹ̥", bytes: [201, 185, 204, 165]
+            // consonant features: {"position":"Alveolar","position_index":5,"manner":"Approximant","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 187:
+        // Phoneme: "ɻ", bytes: [201, 187]
+        // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Approximant","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 203:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 148:
+            // Phoneme: "ɻ˔", bytes: [201, 187, 203, 148]
+            // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Non-sibilant fricative","voiced":true}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 138:
+            // Phoneme: "ɻ̊", bytes: [201, 187, 204, 138]
+            // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Approximant","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 163:
+        // Phoneme: "ɣ", bytes: [201, 163]
+        // consonant features: {"position":"Velar","position_index":9,"manner":"Non-sibilant fricative","voiced":true}
+       return 2;
+          break;
+      case 166:
+        // Phoneme: "ɦ", bytes: [201, 166]
+        // consonant features: {"position":"Glottal","position_index":12,"manner":"Non-sibilant fricative","voiced":true}
+       return 2;
+          break;
+      case 176:
+        // Phoneme: "ɰ", bytes: [201, 176]
+        // consonant features: {"position":"Velar","position_index":9,"manner":"Approximant","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 138:
+            // Phoneme: "ɰ̊", bytes: [201, 176, 204, 138]
+            // consonant features: {"position":"Velar","position_index":9,"manner":"Approximant","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 190:
+        // Phoneme: "ɾ", bytes: [201, 190]
+        // consonant features: {"position":"Alveolar","position_index":5,"manner":"Tap/flap","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 188:
+            // Phoneme: "ɾ̼", bytes: [201, 190, 204, 188]
+            // consonant features: {"position":"Linguo-labial","position_index":3,"manner":"Tap/flap","voiced":true}
+           return 4;
+              break;
+          case 165:
+            // Phoneme: "ɾ̥", bytes: [201, 190, 204, 165]
+            // consonant features: {"position":"Alveolar","position_index":5,"manner":"Tap/flap","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 189:
+        // Phoneme: "ɽ", bytes: [201, 189]
+        // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Tap/flap","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 138:
+            // Phoneme: "ɽ̊", bytes: [201, 189, 204, 138]
+            // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Tap/flap","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 172:
+        // Phoneme: "ɬ", bytes: [201, 172]
+        // consonant features: {"position":"Alveolar","position_index":5,"manner":"Lateral fricative","voiced":false}
+       return 2;
+          break;
+      case 174:
+        // Phoneme: "ɮ", bytes: [201, 174]
+        // consonant features: {"position":"Alveolar","position_index":5,"manner":"Lateral fricative","voiced":true}
+       return 2;
+          break;
+      case 173:
+        // Phoneme: "ɭ", bytes: [201, 173]
+        // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Lateral approximant","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 138:
+            // Phoneme: "ɭ̊", bytes: [201, 173, 204, 138]
+            // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Lateral approximant","voiced":false}
+           if (max_length > 4) {
+            switch(string[cursor + 4]) {
+
+            case 203:
+             if (max_length > 5) {
+              switch(string[cursor + 5]) {
+
+              case 148:
+                // Phoneme: "ɭ̊˔", bytes: [201, 173, 204, 138, 203, 148]
+                // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Lateral fricative","voiced":false}
+               return 6;
+                  break;
+              }
+             } else {
+               return 5;
+             }
+                break;
+              default:
+                return 4;
+            }
+           } else {
+             return 4;
+           }
+              break;
+          case 134:
+            // Phoneme: "ɭ̆", bytes: [201, 173, 204, 134]
+            // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Lateral tap/flap","voiced":true}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+        case 203:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 148:
+            // Phoneme: "ɭ˔", bytes: [201, 173, 203, 148]
+            // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Lateral fricative","voiced":true}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 186:
+        // Phoneme: "ɺ", bytes: [201, 186]
+        // consonant features: {"position":"Alveolar","position_index":5,"manner":"Lateral tap/flap","voiced":true}
+       return 2;
+          break;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 101:
+      // Phoneme: "e", bytes: [101]
+      // vowel features: {"F1":390,"F2":2300,"rounded":false}
+     return 1;
+        break;
+    case 195:
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 184:
+        // Phoneme: "ø", bytes: [195, 184]
+        // vowel features: {"F1":370,"F2":1900,"rounded":true}
+       return 2;
+          break;
+      case 166:
+        // Phoneme: "æ", bytes: [195, 166]
+        // vowel features: {"F1":800,"F2":1900,"rounded":false}
+       return 2;
+          break;
+      case 176:
+        // Phoneme: "ð", bytes: [195, 176]
+        // consonant features: {"position":"Dental","position_index":4,"manner":"Non-sibilant fricative","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 188:
+            // Phoneme: "ð̼", bytes: [195, 176, 204, 188]
+            // consonant features: {"position":"Linguo-labial","position_index":3,"manner":"Non-sibilant fricative","voiced":true}
+           return 4;
+              break;
+          case 160:
+            // Phoneme: "ð̠", bytes: [195, 176, 204, 160]
+            // consonant features: {"position":"Alveolar","position_index":5,"manner":"Non-sibilant fricative","voiced":true}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 167:
+        // Phoneme: "ç", bytes: [195, 167]
+        // consonant features: {"position":"Palatal","position_index":8,"manner":"Non-sibilant fricative","voiced":false}
+       return 2;
+          break;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 197:
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 147:
+        // Phoneme: "œ", bytes: [197, 147]
+        // vowel features: {"F1":585,"F2":1710,"rounded":true}
+       return 2;
+          break;
+      case 139:
+        // Phoneme: "ŋ", bytes: [197, 139]
+        // consonant features: {"position":"Velar","position_index":9,"manner":"Nasal","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 138:
+            // Phoneme: "ŋ̊", bytes: [197, 139, 204, 138]
+            // consonant features: {"position":"Velar","position_index":9,"manner":"Nasal","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 97:
+      // Phoneme: "a", bytes: [97]
+      // vowel features: {"F1":850,"F2":1610,"rounded":false}
+     return 1;
+        break;
+    case 202:
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 140:
+        // Phoneme: "ʌ", bytes: [202, 140]
+        // vowel features: {"F1":600,"F2":1170,"rounded":false}
+       return 2;
+          break;
+      case 138:
+        // Phoneme: "ʊ", bytes: [202, 138]
+        // vowel features: {"F1":350,"F2":650,"rounded":true}
+       return 2;
+          break;
+      case 136:
+        // Phoneme: "ʈ", bytes: [202, 136]
+        // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Stop","voiced":false}
+       return 2;
+          break;
+      case 161:
+        // Phoneme: "ʡ", bytes: [202, 161]
+        // consonant features: {"position":"Pharyngeal","position_index":11,"manner":"Stop","voiced":false}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 134:
+            // Phoneme: "ʡ̆", bytes: [202, 161, 204, 134]
+            // consonant features: {"position":"Pharyngeal","position_index":11,"manner":"Tap/flap","voiced":true}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 148:
+        // Phoneme: "ʔ", bytes: [202, 148]
+        // consonant features: {"position":"Glottal","position_index":12,"manner":"Stop","voiced":false}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 158:
+            // Phoneme: "ʔ̞", bytes: [202, 148, 204, 158]
+            // consonant features: {"position":"Glottal","position_index":12,"manner":"Approximant","voiced":true}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 131:
+        // Phoneme: "ʃ", bytes: [202, 131]
+        // consonant features: {"position":"Post-alveolar","position_index":6,"manner":"Sibilant fricative","voiced":false}
+       return 2;
+          break;
+      case 146:
+        // Phoneme: "ʒ", bytes: [202, 146]
+        // consonant features: {"position":"Post-alveolar","position_index":6,"manner":"Sibilant fricative","voiced":true}
+       return 2;
+          break;
+      case 130:
+        // Phoneme: "ʂ", bytes: [202, 130]
+        // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Sibilant fricative","voiced":false}
+       return 2;
+          break;
+      case 144:
+        // Phoneme: "ʐ", bytes: [202, 144]
+        // consonant features: {"position":"Retro-flex","position_index":7,"manner":"Sibilant fricative","voiced":true}
+       return 2;
+          break;
+      case 145:
+        // Phoneme: "ʑ", bytes: [202, 145]
+        // consonant features: {"position":"Palatal","position_index":8,"manner":"Sibilant fricative","voiced":true}
+       return 2;
+          break;
+      case 157:
+        // Phoneme: "ʝ", bytes: [202, 157]
+        // consonant features: {"position":"Palatal","position_index":8,"manner":"Non-sibilant fricative","voiced":true}
+       return 2;
+          break;
+      case 129:
+        // Phoneme: "ʁ", bytes: [202, 129]
+        // consonant features: {"position":"Uvular","position_index":10,"manner":"Non-sibilant fricative","voiced":true}
+       return 2;
+          break;
+      case 149:
+        // Phoneme: "ʕ", bytes: [202, 149]
+        // consonant features: {"position":"Pharyngeal","position_index":11,"manner":"Non-sibilant fricative","voiced":true}
+       return 2;
+          break;
+      case 139:
+        // Phoneme: "ʋ", bytes: [202, 139]
+        // consonant features: {"position":"Labio-dental","position_index":2,"manner":"Approximant","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 165:
+            // Phoneme: "ʋ̥", bytes: [202, 139, 204, 165]
+            // consonant features: {"position":"Labio-dental","position_index":2,"manner":"Approximant","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 153:
+        // Phoneme: "ʙ", bytes: [202, 153]
+        // consonant features: {"position":"Bi-labial","position_index":1,"manner":"Trill","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 165:
+            // Phoneme: "ʙ̥", bytes: [202, 153, 204, 165]
+            // consonant features: {"position":"Bi-labial","position_index":1,"manner":"Trill","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 128:
+        // Phoneme: "ʀ", bytes: [202, 128]
+        // consonant features: {"position":"Uvular","position_index":10,"manner":"Trill","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 165:
+            // Phoneme: "ʀ̥", bytes: [202, 128, 204, 165]
+            // consonant features: {"position":"Uvular","position_index":10,"manner":"Trill","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 156:
+        // Phoneme: "ʜ", bytes: [202, 156]
+        // consonant features: {"position":"Pharyngeal","position_index":11,"manner":"Trill","voiced":false}
+       return 2;
+          break;
+      case 162:
+        // Phoneme: "ʢ", bytes: [202, 162]
+        // consonant features: {"position":"Pharyngeal","position_index":11,"manner":"Trill","voiced":true}
+       return 2;
+          break;
+      case 142:
+        // Phoneme: "ʎ", bytes: [202, 142]
+        // consonant features: {"position":"Palatal","position_index":8,"manner":"Lateral approximant","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 157:
+            // Phoneme: "ʎ̝", bytes: [202, 142, 204, 157]
+            // consonant features: {"position":"Palatal","position_index":8,"manner":"Lateral fricative","voiced":true}
+           if (max_length > 4) {
+            switch(string[cursor + 4]) {
+
+            case 204:
+             if (max_length > 5) {
+              switch(string[cursor + 5]) {
+
+              case 138:
+                // Phoneme: "ʎ̝̊", bytes: [202, 142, 204, 157, 204, 138]
+                // consonant features: {"position":"Palatal","position_index":8,"manner":"Lateral fricative","voiced":false}
+               return 6;
+                  break;
+              }
+             } else {
+               return 5;
+             }
+                break;
+              default:
+                return 4;
+            }
+           } else {
+             return 4;
+           }
+              break;
+          case 165:
+            // Phoneme: "ʎ̥", bytes: [202, 142, 204, 165]
+            // consonant features: {"position":"Palatal","position_index":8,"manner":"Lateral approximant","voiced":false}
+           return 4;
+              break;
+          case 134:
+            // Phoneme: "ʎ̆", bytes: [202, 142, 204, 134]
+            // consonant features: {"position":"Palatal","position_index":8,"manner":"Lateral tap/flap","voiced":true}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      case 159:
+        // Phoneme: "ʟ", bytes: [202, 159]
+        // consonant features: {"position":"Velar","position_index":9,"manner":"Lateral approximant","voiced":true}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 157:
+            // Phoneme: "ʟ̝", bytes: [202, 159, 204, 157]
+            // consonant features: {"position":"Velar","position_index":9,"manner":"Lateral fricative","voiced":true}
+           if (max_length > 4) {
+            switch(string[cursor + 4]) {
+
+            case 204:
+             if (max_length > 5) {
+              switch(string[cursor + 5]) {
+
+              case 138:
+                // Phoneme: "ʟ̝̊", bytes: [202, 159, 204, 157, 204, 138]
+                // consonant features: {"position":"Velar","position_index":9,"manner":"Lateral fricative","voiced":false}
+               return 6;
+                  break;
+              }
+             } else {
+               return 5;
+             }
+                break;
+              default:
+                return 4;
+            }
+           } else {
+             return 4;
+           }
+              break;
+          case 165:
+            // Phoneme: "ʟ̥", bytes: [202, 159, 204, 165]
+            // consonant features: {"position":"Velar","position_index":9,"manner":"Lateral approximant","voiced":false}
+           return 4;
+              break;
+          case 160:
+            // Phoneme: "ʟ̠", bytes: [202, 159, 204, 160]
+            // consonant features: {"position":"Uvular","position_index":10,"manner":"Lateral approximant","voiced":true}
+           return 4;
+              break;
+          case 134:
+            // Phoneme: "ʟ̆", bytes: [202, 159, 204, 134]
+            // consonant features: {"position":"Velar","position_index":9,"manner":"Lateral tap/flap","voiced":true}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 111:
+      // Phoneme: "o", bytes: [111]
+      // vowel features: {"F1":360,"F2":640,"rounded":true}
+     return 1;
+        break;
+    case 117:
+      // Phoneme: "u", bytes: [117]
+      // vowel features: {"F1":350,"F2":650,"rounded":true}
+     return 1;
+        break;
+    case 109:
+      // Phoneme: "m", bytes: [109]
+      // consonant features: {"position":"Bi-labial","position_index":1,"manner":"Nasal","voiced":true}
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 204:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 165:
+          // Phoneme: "m̥", bytes: [109, 204, 165]
+          // consonant features: {"position":"Bi-labial","position_index":1,"manner":"Nasal","voiced":false}
+         return 3;
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+        default:
+          return 1;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 110:
+      // Phoneme: "n", bytes: [110]
+      // consonant features: {"position":"Alveolar","position_index":5,"manner":"Nasal","voiced":true}
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 204:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 188:
+          // Phoneme: "n̼", bytes: [110, 204, 188]
+          // consonant features: {"position":"Linguo-labial","position_index":3,"manner":"Nasal","voiced":true}
+         return 3;
+            break;
+        case 165:
+          // Phoneme: "n̥", bytes: [110, 204, 165]
+          // consonant features: {"position":"Alveolar","position_index":5,"manner":"Nasal","voiced":false}
+         return 3;
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+        default:
+          return 1;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 112:
+      // Phoneme: "p", bytes: [112]
+      // consonant features: {"position":"Bi-labial","position_index":1,"manner":"Stop","voiced":false}
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 204:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 170:
+          // Phoneme: "p̪", bytes: [112, 204, 170]
+          // consonant features: {"position":"Labio-dental","position_index":2,"manner":"Stop","voiced":false}
+         return 3;
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+        default:
+          return 1;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 98:
+      // Phoneme: "b", bytes: [98]
+      // consonant features: {"position":"Bi-labial","position_index":1,"manner":"Stop","voiced":true}
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 204:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 170:
+          // Phoneme: "b̪", bytes: [98, 204, 170]
+          // consonant features: {"position":"Labio-dental","position_index":2,"manner":"Stop","voiced":true}
+         return 3;
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+        default:
+          return 1;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 116:
+      // Phoneme: "t", bytes: [116]
+      // consonant features: {"position":"Alveolar","position_index":5,"manner":"Stop","voiced":false}
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 204:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 188:
+          // Phoneme: "t̼", bytes: [116, 204, 188]
+          // consonant features: {"position":"Linguo-labial","position_index":3,"manner":"Stop","voiced":false}
+         return 3;
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+        default:
+          return 1;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 100:
+      // Phoneme: "d", bytes: [100]
+      // consonant features: {"position":"Alveolar","position_index":5,"manner":"Stop","voiced":true}
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 204:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 188:
+          // Phoneme: "d̼", bytes: [100, 204, 188]
+          // consonant features: {"position":"Linguo-labial","position_index":3,"manner":"Stop","voiced":true}
+         return 3;
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+        default:
+          return 1;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 99:
+      // Phoneme: "c", bytes: [99]
+      // consonant features: {"position":"Palatal","position_index":8,"manner":"Stop","voiced":false}
+     return 1;
+        break;
+    case 107:
+      // Phoneme: "k", bytes: [107]
+      // consonant features: {"position":"Velar","position_index":9,"manner":"Stop","voiced":false}
+     return 1;
+        break;
+    case 103:
+      // Phoneme: "g", bytes: [103]
+      // consonant features: {"position":"Velar","position_index":9,"manner":"Stop","voiced":true}
+     return 1;
+        break;
+    case 113:
+      // Phoneme: "q", bytes: [113]
+      // consonant features: {"position":"Uvular","position_index":10,"manner":"Stop","voiced":false}
+     return 1;
+        break;
+    case 115:
+      // Phoneme: "s", bytes: [115]
+      // consonant features: {"position":"Alveolar","position_index":5,"manner":"Sibilant fricative","voiced":false}
+     return 1;
+        break;
+    case 122:
+      // Phoneme: "z", bytes: [122]
+      // consonant features: {"position":"Alveolar","position_index":5,"manner":"Sibilant fricative","voiced":true}
+     return 1;
+        break;
+    case 206:
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 178:
+        // Phoneme: "β", bytes: [206, 178]
+        // consonant features: {"position":"Bi-labial","position_index":1,"manner":"Non-sibilant fricative","voiced":true}
+       return 2;
+          break;
+      case 184:
+        // Phoneme: "θ", bytes: [206, 184]
+        // consonant features: {"position":"Dental","position_index":4,"manner":"Non-sibilant fricative","voiced":false}
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 204:
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 188:
+            // Phoneme: "θ̼", bytes: [206, 184, 204, 188]
+            // consonant features: {"position":"Linguo-labial","position_index":3,"manner":"Non-sibilant fricative","voiced":false}
+           return 4;
+              break;
+          case 160:
+            // Phoneme: "θ̠", bytes: [206, 184, 204, 160]
+            // consonant features: {"position":"Alveolar","position_index":5,"manner":"Non-sibilant fricative","voiced":false}
+           return 4;
+              break;
+          }
+         } else {
+           return 3;
+         }
+            break;
+          default:
+            return 2;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 102:
+      // Phoneme: "f", bytes: [102]
+      // consonant features: {"position":"Labio-dental","position_index":2,"manner":"Non-sibilant fricative","voiced":false}
+     return 1;
+        break;
+    case 118:
+      // Phoneme: "v", bytes: [118]
+      // consonant features: {"position":"Labio-dental","position_index":2,"manner":"Non-sibilant fricative","voiced":true}
+     return 1;
+        break;
+    case 120:
+      // Phoneme: "x", bytes: [120]
+      // consonant features: {"position":"Velar","position_index":9,"manner":"Non-sibilant fricative","voiced":false}
+     return 1;
+        break;
+    case 207:
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 135:
+        // Phoneme: "χ", bytes: [207, 135]
+        // consonant features: {"position":"Uvular","position_index":10,"manner":"Non-sibilant fricative","voiced":false}
+       return 2;
+          break;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 196:
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 167:
+        // Phoneme: "ħ", bytes: [196, 167]
+        // consonant features: {"position":"Pharyngeal","position_index":11,"manner":"Non-sibilant fricative","voiced":false}
+       return 2;
+          break;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 104:
+      // Phoneme: "h", bytes: [104]
+      // consonant features: {"position":"Glottal","position_index":12,"manner":"Non-sibilant fricative","voiced":false}
+     return 1;
+        break;
+    case 119:
+      // Phoneme: "w", bytes: [119]
+      // consonant features: {"position":"Labio-velar","position_index":0,"manner":"Approximant","voiced":true}
+     return 1;
+        break;
+    case 106:
+      // Phoneme: "j", bytes: [106]
+      // consonant features: {"position":"Palatal","position_index":8,"manner":"Approximant","voiced":true}
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 204:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 138:
+          // Phoneme: "j̊", bytes: [106, 204, 138]
+          // consonant features: {"position":"Palatal","position_index":8,"manner":"Approximant","voiced":false}
+         return 3;
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+        default:
+          return 1;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 226:
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 177:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 177:
+          // Phoneme: "ⱱ", bytes: [226, 177, 177]
+          // consonant features: {"position":"Labio-dental","position_index":2,"manner":"Tap/flap","voiced":true}
+         if (max_length > 3) {
+          switch(string[cursor + 3]) {
+
+          case 204:
+           if (max_length > 4) {
+            switch(string[cursor + 4]) {
+
+            case 159:
+              // Phoneme: "ⱱ̟", bytes: [226, 177, 177, 204, 159]
+              // consonant features: {"position":"Bi-labial","position_index":1,"manner":"Tap/flap","voiced":true}
+             return 5;
+                break;
+            }
+           } else {
+             return 4;
+           }
+              break;
+            default:
+              return 3;
+          }
+         } else {
+           return 3;
+         }
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 114:
+      // Phoneme: "r", bytes: [114]
+      // consonant features: {"position":"Alveolar","position_index":5,"manner":"Trill","voiced":true}
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 204:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 165:
+          // Phoneme: "r̥", bytes: [114, 204, 165]
+          // consonant features: {"position":"Alveolar","position_index":5,"manner":"Trill","voiced":false}
+         return 3;
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+        default:
+          return 1;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    case 108:
+      // Phoneme: "l", bytes: [108]
+      // consonant features: {"position":"Alveolar","position_index":5,"manner":"Lateral approximant","voiced":true}
+     if (max_length > 1) {
+      switch(string[cursor + 1]) {
+
+      case 204:
+       if (max_length > 2) {
+        switch(string[cursor + 2]) {
+
+        case 165:
+          // Phoneme: "l̥", bytes: [108, 204, 165]
+          // consonant features: {"position":"Alveolar","position_index":5,"manner":"Lateral approximant","voiced":false}
+         return 3;
+            break;
+        }
+       } else {
+         return 2;
+       }
+          break;
+        default:
+          return 1;
+      }
+     } else {
+       return 1;
+     }
+        break;
+    }
+  return 0;
+}