# Changelog of Whisper Transcriber — Video Links & Files to Text (`lergassy/whisper-transcriber`) Actor

- **URL**: https://apify.com/lergassy/whisper-transcriber/changelog.md
- **Full Actor documentation**: https://apify.com/lergassy/whisper-transcriber.md

## Changelog

### 2026-10-02

- Default memory now scales with your input, so small runs start faster and use less memory.

### 2026-10-01

- Video links are checked for length and size before any audio is downloaded; a file that does not fit your maximum charge is skipped with a clear row saying how many minutes it needs.
- Video pages now fetch a lighter audio-only track that carries everything recognition needs, so downloads are smaller and quicker.
- Pages that only serve a very heavy video stream are declined up front with a free row suggesting a direct file link instead.

### 2026-09-27

- Runs stop cleanly at your maximum cost instead of being aborted; nothing beyond your limit is written or charged.

### 2026-09-26

- SRT and WebVTT subtitles now follow the usual reading norm: at most two lines of 42 characters per caption and no caption longer than 7 seconds.

### 2026-09-24

- Files on hosts that use redirects or short links are retried with a second download method before being reported as unreachable.
- An empty or invalid input now returns a free row explaining what to fix instead of a failed run.

### 2026-09-19

- The input form now lists the 14 supported video sites, and each example input shows a different use case.
- Video pages download audio only, capped at 150 MB; anything larger returns a free explanation row.

### 2026-09-11

- Initial release: transcribe audio and video files or video links with Whisper large v3 in 90+ languages, with timecodes, optional speaker labels, English translation, SRT/WebVTT subtitles and RAG-ready chunks.