Skip to content

Babelin Speach, for voice recognition and real-time translation, services offered by web browsers

Notifications You must be signed in to change notification settings

caixiaotian1989/Babelin

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 

Repository files navigation

Babelin Speach

It is a simple html file for voice recognition and Real-time translationof files videos or audio, use the services offered by web browsers (Chrome/Edge current). It is possible to see the subtitles in real time (delay minus 1 second), in any video and in any language that the browser supports.

IMAGE ALT TEXT HERE

Features⚙️

  • Transcription and translation in real time.
  • Creation subtitles.
  • Allows you to set the time to divide sentences.
  • It allows to edit the subtitles using the generated table.
  • Import and export in three subtitle formats vtt, ass, srt.
  • Allows to join/splits adjacent lines
  • Shortcuts for time correction, recognition start and stop, forward and backward to the next subtitle cue, video advance in 500ms, 1sec, and 10sec.
  • Option to translate the text with other translators such as DeepL, Yandex, Bing. With copy and paste, and a word counter to facilitate this.
  • Time correction for recognition delay.
  • Create line new empty press key c.

Requirements 📋

Installation 🔧

Built with 🛠️

Download 📌

V 1.0.

Test online with the latest version 1.1

All audio is sent to google/cloud servers for processing.

Version 1.1

  • Colors are added to the table based on CPS.
  • Button "Fix Time Start" lines short time.
  • Add Button for Download Table or Track of subtitle.
  • Compatible Edge/Chrome
  • By default, groups of statements/sentence/phrases are handled.
  • Many bugs are fixed

*CPS stands for "characters per second". It therefore measures the speed at which words appear in subtitles or captions.

Recommendations

  • Use Edge/Microsoft performs better transcription and Chrome/Google for It better recognizes the start and end times of the voice, but its transcription is regular.

  • There are two important CPSs, the source CPS and the destination CPS. For example, languages like Japanese, Korean, Chinese can have CPS of 4-6, while languages like English 8-12, or Portugues/Spanish of 10-14.

  • Once the whole transcription is finished, if colored lines appear, easily by pressing the "Fix Time Start (CPS)" button, correct the start times of the lines with short time problems.

Shortcuts General

  • Ctrl+Q: Start/Stop Recognition Speech
  • alt + PageDown: Next cue
  • alt + PageUp: Previous cue

SHORTCUTS IN TABLE

  • Ctrl+alt+ArrowUp: Join row previous
  • Ctrl+alt+ArrowDown: Join row next
  • Ctrl+alt+J: Split row in cursor or selection

SHORTCUTS IN CELLS TIME

  • minus (-): rest 100 ms to start time
  • plus (+): add 100 ms to start time

SHORTCUTS IN VIDEO

  • Q: Start/Stop Recognition Speech

  • P: Play/Stop video

  • C: Hold down the key C, create a new subtitle line empty

  • Alt + ArrowUp: Forward 5 seconds

  • Alt + ArrowDown: Rewind 5 seconds

  • Ctrl + ArrowUp: Forward 10 seconds

  • Ctrl + ArrowDown: Rewind 10 seconds

  • PageDown: Next cue

  • PageUp: Previous cue

  • M: Forward 1 second

  • Shift + M: Forward 500 ms

  • N: Rewind 1 second

  • Shift + N: Rewind 500 ms

  • Del: Delete cue(s) active

  • Ctrl + Del: Delete all cue(s) active

  • +: add 100 ms to start time of last cue active.

  • -: rest 100 ms to star time of last cue active.

  • Alt + +: add 100 ms to start time of last cue active(All).

  • Alt + -: rest 100 ms to star time of last cue active(All).

  • ctrl + +: add 100 ms to end time of last cue active.

  • ctrl + -: rest 100 ms to end time of last cue active.

Subtitles and CPS

https://captiz.com/wp-content/themes/captiz/SubtitlingGuidelinesCaptiz.pdf

https://partnerhelp.netflixstudios.com/hc/en-us/articles/215758617-Timed-Text-Style-Guide-General-Requirements

https://www.authot.com/en/2022/02/10/synchronisation-of-subtitles/

About

Babelin Speach, for voice recognition and real-time translation, services offered by web browsers

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • HTML 100.0%