markshust

<template>
  <div>
    <input
      type="text"
      placeholder="Search Your Interest"
      @input="debounceSearch()"
      v-model="searchInput"
    />
  </div>
</template>

markshust

Overview

I just read this trick for text compression, in order to save tokens in subbsequent interactions during a long conversation, or in a subsequent long text to summarize.

SHORT VERSION:

It's useful to give a mapping between common words (or phrases) in a given long text that one intends to pass later. Then pass that long text to gpt-4 but encoded with such mapping. The idea is that the encoded version contains less tokens than the original text. There are several algorithms to identify frequent words or phrases inside a given text, such as NER, TF-IDF, part-of-speech (POS) tagging, etc.

markshust

#!/bin/bash

languageName='Spanish'
language='es'

for file in *.srt; do
  if [[ $file == *.$language.srt ]]; then
    # Skip processing of already translated files
    continue
  fi

markshust

import os
import whisper
from whisper.utils import get_writer

# Get the current directory path
directory = os.getcwd()

# Loop through all the files in the directory
for file in sorted(os.listdir(directory)):
  

markshust

#!/bin/bash

# Requires translate-shell to be installed.
# Get it on GitHub at: https://github.com/soimort/translate-shell

# Parse command-line arguments
while [[ "$#" -gt 0 ]]; do
    case $1 in
        --file|-f ) file="$2"; shift ;;
        --language|-l ) language="$2"; shift ;;

markshust

import os
import textwrap

input_dir = 'input1/'
output_dir = 'output1/'

for filename in sorted(os.listdir(input_dir)):
    if filename.endswith('.srt'):
        with open(input_dir + filename, 'r') as input_file:
            output_filename = output_dir + filename

markshust

import os
import whisper
from whisper.utils import get_writer

input_dir = 'input/'
output_dir = 'output/'

for filename in sorted(os.listdir(input_dir)):
    if filename.endswith('.mp4'):
        input_file = input_dir + filename

markshust

content

markshust

#!/bin/bash

course="php-101"

totalcount=0
for i in *.mp4; do
  [[ $i == *"- clean"* ]] && continue
  totalcount=$((totalcount+1))
done;

markshust

#!/bin/bash

for i in *.mp4; do
  [[ $i == *"- clean"* ]] && continue # this line excludes files that contain "- clean" in filename
  prefix=`echo "$i" | sed "s/^\([0-9]*\)\.\(.*\)/\1/"`;
  ffmpeg -i "$i" -ss 00:00:15.00 -vf "thumbnail,scale=480:270" -frames:v 1 "$prefix.webp"
done