phonemenon/followerfreq-analysis-stream.js at master · jimkang/phonemenon · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
// It's a stream that analyzes phoneme groups you write to it, then returns the
// analysis via the callback when the end is reached.

var Writable = require('stream').Writable;
var pluck = require('lodash.pluck');

function createAnalyzeFollowerStream(opts) {
  var followerFreqsForPhonemes = {};

  // Must use object mode.
  if (!opts.objectMode) {
    opts.objectMode = true;
  }

  var analyzeStream = new Writable(opts);

  analyzeStream._write = function writeToStream(group, encoding, callback) {
    recordPhonemeGroupFollowFrequencies(group);
    callback();
  };

  analyzeStream.end = function end() {
    if (opts.done) {
      opts.done(null, followerFreqsForPhonemes);
    }
  };

  // Incoming phoneme groups will look like this:
  // {
  //   "word": "ABALONE",
  //   "phonemes": [
  //     {"phoneme":"AE","stress":2},
  //     {"phoneme":"B","stress":-1},
  //     {"phoneme":"AH","stress":0},
  //     {"phoneme":"L","stress":-1},
  //     {"phoneme":"OW","stress":1},
  //     {"phoneme":"N","stress":-1},
  //     {"phoneme":"IY","stress":0}
  //   ]
  //
  // }
  //
  // They may also have this optional syllables array:
  //   "syllables": [
  //     [
  //       "AE"
  //     ],
  //     [
  //       "B",
  //       "AH"
  //     ],
  //     [
  //       "L",
  //       "OW"
  //     ],
  //     [
  //       "N",
  //       "IY"
  //     ]
  //   ]

  function recordPhonemeGroupFollowFrequencies(group) {
    var phonemeSequence;
    if (opts.analyzeInSyllables) {
      group.syllables.forEach(recordPhonemeSequenceFrequencies);
    } else {
      phonemeSequence = pluck(group.phonemes, 'phoneme');
      recordPhonemeSequenceFrequencies(phonemeSequence);
    }
  }

  function recordPhonemeSequenceFrequencies(sequence) {
    var phonemeSequence;
    var previousPhoneme;

    if (opts.reverse) {
      phonemeSequence = ['START'].concat(sequence);
      previousPhoneme = 'END';
    } else {
      phonemeSequence = sequence.concat('END');
      previousPhoneme = 'START';
    }

    for (var i = 0; i < phonemeSequence.length; ++i) {
      var phoneme;
      if (opts.reverse) {
        phoneme = phonemeSequence[phonemeSequence.length - 1 - i];
      } else {
        phoneme = phonemeSequence[i];
      }

      if (previousPhoneme) {
        var freqs = {};
        if (previousPhoneme in followerFreqsForPhonemes) {
          freqs = followerFreqsForPhonemes[previousPhoneme];
        } else {
          followerFreqsForPhonemes[previousPhoneme] = freqs;
        }

        var frequency = 0;
        if (phoneme in freqs) {
          frequency = freqs[phoneme];
        }
        frequency += 1;
        freqs[phoneme] = frequency;
      }
      previousPhoneme = phoneme;
    }
  }

  return analyzeStream;
}

module.exports = createAnalyzeFollowerStream;