Workflow directory

Private AI audio tools for the exact job you need to do

The tools directory gathers cleanup utilities, speech-to-text, subtitle, creator voice, and Ebook & Document Audio workflows in one place. Use it when you know the task you need, not just the model you want.

15 focused tools 4 workflow families Free to use · no signup or API key

01 · Prepare

Tools

Clean text, fix punctuation, strip subtitle noise, and prepare scripts before generation or export.

4 tools

🧹

Text Cleaner

Clean text for TTS, content creation, and data processing

✏️

Punctuation Enhancer

Fix punctuation and add pauses for better TTS output

📝

Subtitle Cleaner

Strip timestamps and extract clean text from SRT and VTT files

🎬

Script Formatter

Format scripts for natural-sounding TTS output

Transcription & Subtitles

Private browser workflows for speech to text, captions, and subtitle exports.

5 tools

📝

Audio to Text

Convert audio or video into private transcripts and subtitle exports

📝

Video to Text

Transcribe a video file into reviewable text, captions, and subtitle exports

📝

MP3 to Text

Convert MP3 recordings into private transcripts and editable subtitle files

🎬

Subtitle Generator

Create caption drafts and download SRT or VTT subtitle files from media you select

📝

SRT Generator

Generate an editable SRT subtitle draft with timestamps from audio or video

03 · Create

Creator Voice Tools

Voice-over workflows for creators who want browser-based narration and repurposing assets.

3 tools

▶️

YouTube Voice Generator

Generate browser-based voice-over for YouTube scripts and narration

▶️

Free AI Voice Generator

Turn text into downloadable speech with browser-based AI voices

🇫🇷

French Pronunciation Generator

French Pronunciation Generator for listening to and practicing French text

04 · Listen

Ebook & Document Audio

Turn EPUB, PDF, and TXT reading material into speech for accessibility, review, and long-form listening.

3 tools

📚

Ebook to Audio

Convert EPUB, PDF, and TXT reading material into speech in your browser

📚

PDF to Audio

Listen to text-based PDF documents with browser-based AI voices

📚

EPUB to Audiobook

Turn EPUB chapters into downloadable audiobook-style listening drafts

Choose the Tool by the Output You Need

These pages are not interchangeable keyword versions of one generator. Each one opens a different component or a focused view of a shared engine. Start from the artifact you need to leave with, then check the limits before moving sensitive or production material into the workflow.

Needed output	Start here	What it changes	Review before use
Plain text without web or formatting noise	Text Cleaner	Removes selected URLs, email addresses, HTML, emoji, invisible characters, or extra spacing	Important symbols and contact details may be removed
A narration draft with more deliberate punctuation	Punctuation Enhancer	Applies deterministic punctuation and capitalization rules	Rules can misread names, dialogue, code, or specialist text
Spoken lines extracted from SRT or VTT	Subtitle Cleaner	Strips selected cue notation, labels, sounds, tags, and duplicates	SRT/VTT output uses placeholder timing and needs retiming
A structured creator or audiobook script	Script Formatter	Adds line breaks, planning markers, pauses, and an approximate duration	Labels are not speaker detection and may be read aloud
TXT transcript or SRT/VTT captions	Audio to Text	Runs Whisper transcription with segment or optional word timing	Names, numbers, speakers, and timestamps need human review
Narration audio for a video timeline	YouTube Voice Generator	Turns a reviewed script into downloadable speech	It does not create video, clone a person, publish, or guarantee monetization
A resumable book with paragraph audio	Ebook to Audio	Parses supported files, stores a local bookshelf, and generates selected chapters	Scanned PDFs need OCR; storage and phonemization boundaries apply

What “Private Browser Workflow” Means

Text Cleaner, Punctuation Enhancer, Subtitle Cleaner, and Script Formatter apply their rules in page JavaScript. Audio to Text runs Whisper inference on the selected media in the browser. TTS engines synthesize audio locally after their assets load. This avoids an upload-first account dashboard, but it does not mean the whole website is disconnected from the network.

The site, analytics, and model hosts receive ordinary web requests. Non-English Kokoro sends entered text to the documented phonemization service before local synthesis. First-time model use needs a download, and browser storage may retain models, preferences, ebook text, or generated paragraph audio until cleared. The Privacy Policy explains these paths and deletion options.

A Safer Production Order

Keep an unchanged source copy. Work on a short representative sample, select only the cleanup rules you need, compare the output, and then test the final TTS or STT engine. Check names, numbers, quotations, meaningful sound labels, and rights-sensitive material before exporting. Local processing does not make an inaccurate transcript correct or grant permission to reuse someone else's book, captions, recording, or voice identity.

For production delivery, listen to the complete audio or compare the full transcript with the source. Save the reviewed text alongside the exported file so later corrections are possible. Clear site data when local retention is inappropriate, remembering that exported files must be deleted separately from the operating system. Report reproducible tool failures through the Contact page with private content removed.

Need the raw engine workspace instead?

Go straight to the TTS or Whisper workspace when you already know which engine-driven tool you want.

Open TTS Workspace → Open STT Workspace