# Changelog of Transcribe Video to Text & Audio to Text — 99+ Languages (`sian.agency/incredibly-fast-audio-transcriber`) Actor

- **URL**: https://apify.com/sian.agency/incredibly-fast-audio-transcriber/changelog.md
- **Full Actor documentation**: https://apify.com/sian.agency/incredibly-fast-audio-transcriber.md

## Changelog

All notable changes to **Transcribe Audio & Video to Text — 99+ Languages** will be documented in this file.

### 2026-10-03

- Processing reports now include copyable transcripts and subtitle text, accepted per-second events and the platform start fee when runtime billing data is available.
- Output schemas describe word/segment timings and nullable speaker, language and source fields.
- Service error rows show HTTP status and retry guidance without raw service response bodies.
- Input and documentation now explain the sample URL, caption timing limits, current-price lookup and Apify storage retention.

### 2026-05-21 (later)

#### 💎 YouTube / TikTok / Instagram URLs are now a PAID-tier feature

- **Platform URL Transcription is PAID** — YouTube, TikTok, and Instagram links are unlocked on the PAID tier. The FREE tier continues to support direct media file URLs (MP3, WAV, MP4, MOV, M4A, OPUS, …) and uploaded files.
- **Clear FREE-Tier Rejection Message** — FREE-tier users who paste a platform URL get an immediate, dataset-visible explanation and an upgrade path, with **zero result charges** for the rejected URL (the platform start fee is separate).
- **No Wasted Calls** — the gate runs **before** any platform extraction or transcription, so FREE-tier accounts can never accidentally trigger an upstream call against a YouTube/TikTok/Instagram URL.

#### 💎 User Benefits

- Transparent tier boundaries — FREE-tier users know up front what works and what requires an upgrade
- PAID users get the full multi-platform experience: direct files + YouTube + TikTok + Instagram in one bulk run
- No transcription result charges for rejected FREE-tier platform URLs

#### 🎯 Use Cases

- **Marketing teams** on the PAID tier transcribing trending TikToks, Instagram Reels, and YouTube videos for content research
- **Agencies** running large-scale platform-URL transcription jobs without worrying about FREE-tier limits

***

### 2026-05-21

#### 🚀 Paste YouTube, TikTok & Instagram Links — Transcribe Anything in One Run

- **Native YouTube Transcription** — paste any YouTube URL into `audioUrls` and get a clean text transcript back, no manual download step
- **TikTok Transcription Built In** — drop in a TikTok video URL and the actor extracts the audio and transcribes it for you
- **Instagram Reel & Post Transcription** — paste Instagram Reel or post URLs alongside your other inputs; processed in the same bulk run
- **Source Platform Tracking** — every dataset row now ships with a `sourcePlatform` field (`direct` · `youtube` · `tiktok` · `instagram`) for easy filtering and analytics
- **Resolved Media URL** — new `mediaUrl` field exposes the underlying media stream that was transcribed, so you can audit exactly what the engine processed
- **Lower Latency on Direct Files** — direct media URLs now transcribe faster end-to-end
- **Reworked FREE Tier** — 1 URL/file per run, capped at ≤5 MB or ≤60 seconds, with a hard precheck **before** any transcription cost

#### 💎 User Benefits

- Skip the "download the file first" detour for YouTube, TikTok, and Instagram content
- Run a single bulk job that mixes direct files, uploads, and platform URLs together
- Audit exactly what was transcribed for every row, even when the input was a platform link
- Never get surprise-charged on the FREE tier — over-cap files are rejected before transcription

#### 🎯 Use Cases

- **Podcasters** transcribing both the direct MP3 and the YouTube version of every episode in one run
- **Social-media editors** turning trending TikToks and Reels into searchable text for content research
- **Journalists & researchers** pulling clean transcripts from Instagram interviews and YouTube panel discussions

#### 📝 Good to Know

- For YouTube URLs, word-level timestamps and speaker labels are not available (captions-only routing). Use a direct file URL when you need per-word timing or diarization on YouTube content.

***

### 2026-04-27

#### 📤 Direct File Upload + SRT/VTT Subtitles + Word-Level Timestamps

- **Upload Audio & Video Files Directly** — drop media into the new `audioFiles` field straight from your computer; no need to host files first
- **Mix URLs and Uploads in One Run** — pasted URLs and uploaded files are processed together with the same dedup, validation, and tier limits
- **SRT and VTT Subtitle Output** — every successful transcription now ships with ready-to-use `srt` and `vtt` fields. Save as `.srt` / `.vtt` and drop into any video player or HTML5 `<track>` element
- **Word-Level Timestamps** — `segments[].words[]` now exposes `{word, start, end, speaker}` for every spoken word. Build karaoke-style highlighting, precise quote search, or word-accurate clipping
- **Restructured README** — new "How to transcribe", "Example output", "Speaker diarization", and "SRT / VTT subtitle export" sections lead the page so you can see what the actor does at a glance
- **Clearer Title & Description** — rewritten to plainly describe what the tool does (transcribe video and audio to text, 99+ languages, speaker diarization, SRT/VTT export)

#### 💎 User Benefits

- Skip the "host the file somewhere first" detour — upload `.m4a`, `.mp3`, `.mp4`, etc. straight from your computer
- Get subtitle files ready to publish without any post-processing
- Build searchable, clip-accurate transcripts thanks to word-level timing
- Mix URL inputs and direct uploads in a single bulk run

#### 🎯 Use Cases

- **Podcasters** publishing show notes alongside each new episode
- **Sales and ops teams** archiving meeting recordings for coaching and QA
- **Journalists** turning phone-recorded interviews into clean attributed transcripts
- **Video editors** generating accurate caption tracks for long-form content
- **Students and educators** transcribing lectures and study sessions

***

### 2026-04-07

#### 🛡️ Smart URL Validation

- **Social Media URL Detection** — Facebook, X/Twitter, and Vimeo links are automatically rejected with helpful redirect messages to platform-specific actors
- **File Type Validation** — non-media file extensions (HTML, images, PDFs) are blocked before processing, saving runtime and credits
- **Actor Recommendations** — users who paste social-media links get direct links to the correct SIÁN actor for their platform
- **Graceful Handling** — invalid URLs are reported in the dataset with clear error messages; valid URLs in the same batch still get processed
- **Updated Input Schema** — prominent warnings and valid/invalid URL examples directly in the Apify UI

***

### 2025-11-18

#### 🔗 Smart URL Handling

- **Auto-URL Redirect** — paste any audio URL format, the actor handles the rest
- **Zero URL Hassle** — no more manual conversion to direct download links
- **Universal URL Support** — works with raw URLs from major hosting providers
- **Seamless Experience** — just paste and transcribe

***

### 2025-11-08

#### 🎉 Transcribe Audio & Video to Text — Launch!

- **10× Parallel Processing** — concurrent file handling (10 files simultaneously) on the paid tier
- **Paid tier processing** — up to 10 concurrent files
- **Bulk processing** — actual processing time depends on file length and service load
- **1 GB File Support** on the paid tier — handle long-form audio and video without splitting
- **Unlimited Monthly Volume** for the paid tier
- **SIÁN Agency Branding** — professional store presence and consistent quality across the actor portfolio

#### 💎 User Benefits

- Run production-scale bulk transcription jobs in a fraction of the time
- Transcribe long-form audio and video without splitting or pre-processing
- Audio processing is billed per second; the platform start fee and enabled add-ons are separate

#### 🎯 Use Cases

- **Podcast networks** transcribing entire back catalogs in a single overnight job
- **Research teams** turning hundreds of interview recordings into searchable text
- **Course creators** auto-captioning their full video libraries for accessibility