jimregan

1
00:00:01,387 --> 00:00:03,387
[mwahahahaha]

2
00:00:04,202 --> 00:00:11,662
So... Phoneme recognition for fan and prophet: uses of more direct represations- representations of speech.

3
00:00:13,620 --> 00:00:15,840

jimregan

jimregan

#!/usr/bin/perl
use warnings;
use strict;
use utf8;

binmode(STDIN, ":utf8");
binmode(STDOUT, ":utf8");
binmode(STDERR, ":utf8");

my %skip = map { $_ => 1 } qw(lj_d ll_d m_d nj_d nn_d r_d rj_d);

jimregan

jimregan

# VAD wrapper is taken from PyTorch Speaker Verification:
# https://github.com/HarryVolek/PyTorch_Speaker_Verification
# Copyright (c) 2019, HarryVolek
# License: BSD-3-Clause
# based on https://github.com/wiseman/py-webrtcvad/blob/master/example.py
# Copyright (c) 2016 John Wiseman
# License: MIT
# Additions Copyright (c) 2021, Jim O'Regan
# License: MIT
import collections

jimregan

Based on
https://www.dyslexia-reading-well.com/44-phonemes-in-english.html
adapted for US English, and with additions

Lexicon based on wiktionary (but normalised)

jimregan

Lexicon is based on the one from https://github.com/kylebgorman/wikipron
wikipron has all IPA symbols split; I've rejoined (what I assume are) diphthongs

Alignables file was made based on the lexicon; there are undoubtedly missing alignables

Needs festus (part of https://github.com/google/language-resources/)

jimregan

https://en.wikipedia.org/wiki/Manx_language

jimregan

/* print middle element of list; second if even */
/* http://practice.geeksforgeeks.org/problems/finding-middle-element-in-a-linked-list/1 */
/* tl;dr - two pointers, one incrementing by two  */

#include <stdio.h>
#include <stdlib.h>

struct Node {
        int data;
        struct Node* next;

jimregan

    private static String dropSpaces(String s) {
        String out = "";
        for(char c : s.toCharArray()) {
            if(c != ' ') {
                out += c;
            }
        }
        return out;
    }
    public static String stringDropsString(String haystack, String needle) {