Dos

Audio transcription and speech to text, Whisper included — no API key. Audio/video files or podcast feeds to text, SRT + VTT, 99+ languages. Pay per minute; silent or failed files never billed. Built on the open-source Whisper model (MIT licence); not affiliated with or endorsed by OpenAI.

Transcribe podcast episodes straight from an RSS feed. Pick the newest N per show, only episodes after a date, or only titles matching a phrase. Whisper runs inside the Actor — no API key. Show and episode metadata, text, SRT + VTT. Failed or silent episodes are never billed.

Turn scanned or image-only PDFs into structured text that keeps its reading order: per-block bounding boxes, per-block OCR confidence, column-aware ordering, and a spatial plain-text rendering. Arabic and English OCR built in. Pay per document plus per page.

Greenhouse, Lever, Workday, SmartRecruiters, Recruitee and Workable job scraper — open roles pulled straight from company career sites across 6 ATS platforms into one clean JSON schema. Salary where published, no recruiter names or emails, no logins or proxies. Pay per job returned.

Clean, deduplicate, and type-infer CSV, Excel, and JSON files from URLs. Returns a quality report, preview rows, and a download link to the cleaned CSV. Built for AI agents and data pipelines.

Convert DOCX files, PDFs, and web pages into clean markdown for LLM and agent ingestion. Main-content extraction strips navigation junk from web pages. Pay per document.

Extract every table from PDF files into clean, header-mapped JSON rows. Built for AI agents, data pipelines, and spreadsheet workflows. Pay per document processed.

Split any recording into speaker turns: transcript per speaker, talk-time stats, speaker-labelled SRT/VTT and ready-to-paste Markdown meeting notes. Whisper + voice clustering run inside the Actor — no API key. Failed or silent files are never billed.

Turn audio or video into ready-to-use .srt and .vtt subtitle files. Whisper runs inside the Actor — no API key. Cues are re-segmented to your characters-per-line, lines-per-cue and max-duration limits, with a sync offset. 99+ languages. Failed or silent files are never billed.

Extract text from images with Tesseract inside the Actor — no API key. Plain text plus per-line bounding boxes and confidence, auto script detection and rotation fix, Arabic/Latin/Cyrillic and more. Only images that deliver text are billed.

Extract clean plain text from PDF files: whole document plus every page separately, in true reading order — multi-column layouts and RTL (Arabic) handled. No OCR overhead, no API keys. Pay per page delivered; scanned pages are detected, reported and never billed.