Changelog

ScribeToAny in ChatGPT and Claude

• Connect your account to AI apps: Add ScribeToAny to ChatGPT, Claude or another app that supports MCP, then work with your transcripts right in the chat. See how to connect.
• Ask for transcripts: "List my recent transcripts", or "Get the transcript of the budget meeting with timestamps". You get plain text, or SRT and VTT subtitles.
• Transcribe from the chat: Share a link from X, Twitch, Loom, Google Drive or Dropbox. In ChatGPT you can also attach an audio or video file; in other apps, ask for an upload link.
• You stay in control: You sign in and approve each app. Apps can read your transcripts and start new transcriptions, but they can't edit or delete anything. Transcriptions started from an AI app count toward your plan as usual.

Export to Notion

• One click to Notion: Choose Export to Notion on any finished transcript. It becomes a new private page in your Notion workspace.
• Ready to read: Every line keeps its timestamp and speaker label, with the language, the duration and a link back to the recording at the top. Recordings of several hours fit on one page.
• Connect your workspace the first time you export, and disconnect at any time from the same panel.

Transcription

• Files up to 6 hours: Paid plans and minute packs now take files up to 6 hours long, up from 3. Speaker detection works across the whole file.
• All 99 languages: The language picker lists every language the engine supports, with native names, and you can search it by typing.
• A heads-up when the language is unclear: If auto-detect isn't sure which language was spoken, the transcript page says so, so you can pick the language and transcribe again.
• Easier-to-read subtitles: Each subtitle now stays on screen a moment after the words end, the way streaming services time them, without overlapping the next line.
• More reliable long recordings: Fewer repeated lines in long or quiet recordings, faster imports of long X (Twitter) videos, and a job that would go over your remaining minutes is stopped before the download starts, not after.

AI summary & chat

• Summary: Open AI summary & chat on any finished transcript to get the key points, decisions and action items. Every point links to the moment it was said, so one click jumps the player there.
• Chat: Ask questions about the recording, such as "What did they decide about the budget?" Answers cite timestamps too, and the conversation is kept with the transcript.
• Long recordings: Recordings of several hours are summarized part by part, then pulled together into one summary.
• Available on every plan: the Free plan has a small daily limit, paid plans a higher one. It's optional: a transcript's text goes to a third-party AI provider only when you use it.

Minute packs

• One-time minute packs: 300, 800 or 2,000 minutes for $5, $10 or $20. They're a single payment that never renews, and they're valid for 12 months.
• Spent after your plan's minutes: Packs are used only once your monthly minutes run out. On the Free plan, holding a pack also unlocks Plus limits for file length, files per day and parallel transcriptions.
• Usage alerts: We email you when you've used 80% and 100% of your monthly minutes, so a long recording never stops you by surprise.
• Clearer limits on Free: When you pick a file longer than 30 minutes, the transcribe dialog tells you before uploading that only the first 30 minutes will be transcribed.

Free subtitle translator

• Translate subtitle files without an account: On the translation tool pages, upload an SRT, VTT or other subtitle file, translate it with Google Translate, and bring the result back to edit it.
• Bilingual export: Download the original, the translation, or both lines together, in 8 formats.
• Better file reading: SRT files without blank lines between cues, ASS styling tags and older text encodings are now read correctly.

Translation

• Free translation into 134+ languages: Click Translate on any finished transcript. It opens in Google Translate, where you pick one of 134+ languages. Click Save to ScribeToAny subtitles and the translation comes back as its own language track. Available on every plan, including Free.
• Original timings kept: Only the text is translated. Every line keeps its start and end time, so a translated SRT or VTT lines up with your video exactly like the original.
• Switch and export per language: Buttons above the transcript switch between Original and each saved language. Export gives you whichever version is on screen. Prefer files only? Download the translated TXT, SRT, VTT or CSV straight from the translation page.
• This replaces the 56-language translation introduced in v2.1: more languages, and no longer tied to a plan. See how to translate subtitles, step by step.

Editing

• QC mode: A subtitle-review view for the transcript, with optional reading-speed and line-length checks.
• Version history: A version is saved before each editing session in QC mode, and you can roll the transcript back to any of them.

Transcription

• Number of speakers: If you know how many people are talking, set it (2–8) in the transcribe dialog. Otherwise leave it on auto-detect.

Plans

• New Plus plan: A new Plus plan now sits between Free and Max, so there's a middle option for regular use. See what each plan includes on the pricing page.
• Yearly billing up front: Plans now show yearly pricing by default, so the annual saving is visible at a glance. Monthly is still one click away.

Transcription Quality

• Stronger speaker recognition: Upgraded diarization gives cleaner speaker labels and stays stable on long recordings of several hours.

Exports

• Branded PDF & DOCX: Human-facing PDF and DOCX exports now carry a small "Transcribed with ScribeToAny" footer with a link. Subtitle and data formats (SRT, VTT, TSV, CSV, JSON) are untouched.

Trust & Help

• New pages: A Security page explaining how your files are protected, a Help Center, and a clear Refund & Cancellation policy.
• Refreshed emails: Transactional emails (verification, notifications) now use a cleaner, branded layout.

A new suite of free tools that run entirely in your browser — no upload, no account, no cost. Your files never leave your device.

Subtitle Tools

• Subtitle converter: Convert between SRT, VTT, TXT, CSV and JSON, and export to DOCX or PDF.
• Subtitle editor: Fix cues and re-time captions in the browser, then export again.

Media Utilities

• Extract & convert audio: Pull the audio out of a video, or convert between audio formats.
• Video conversions: MP4 → WebM and video → GIF.
• Image conversion: HEIC → JPG.

How It Works

All of these run locally with WebCodecs and WebAssembly — nothing is uploaded to a server. They are separate from transcription and completely free. Browse them at /tools.

Translation

• 56 target languages: Pick a target language when you start a job and the finished transcript is translated into it — Simplified and Traditional Chinese, Spanish, French, German, Japanese, Korean, Arabic, Hindi and 48 more. The list is the set the model was actually trained and benchmarked on, not every code it will nominally accept.
• Timings are preserved: Translation runs segment by segment and each result is written back against the timestamp it came from. The translated SRT or VTT drops straight into your editor and stays in sync.
• Both documents are kept: The original transcript and the translation live side by side under the same job, and either can be exported in all eight formats. The target language is picked when the job starts and can't be changed afterwards.

Availability

Translation is part of the Max plan. Transcribing non-English audio directly into English remains available on every plan, including the free one.

New Features

Following extensive feedback from creators, v2.0 brings several major efficiency upgrades:

• Speaker Recognition: Interviews and meetings are labelled by speaker, and cues break when the speaker changes instead of mid-answer — so a two-person conversation reads like one.
• Import From a Link: Paste a direct media URL and we fetch and transcribe it, no download-then-upload round trip. (YouTube links are not supported — download the file first.)
• Eight Export Formats: Beyond plain text, export precisely timed SRT and VTT subtitles, spreadsheet-ready TSV and CSV, structured JSON, and print-ready PDF and DOCX. Pick several at once and you get a ZIP.

UX and Performance Improvements

• Resumable Large Uploads: A newly designed upload component now supports massive video files up to 5GB with automatic retries on unstable networks.
• Dashboard Upgrades: Redesigned UI with transcript history search, folders for organising jobs, and batch export.
• Enhanced Noise Resistance: Optimized voice isolation algorithms specifically tailored for outdoor interviews and noisy cafe environments.

Core Features

We are incredibly excited to announce the first official release of ScribeToAny! The intelligent transcription tool built for creators and professional teams is now available:

• Top-Tier Engine: Powered by the industry-leading Whisper large model, delivering unparalleled recognition accuracy.
• Universal Format Support: Directly upload popular audio/video formats including MP3, WAV, FLAC, MP4, and MOV.
• Blazing Fast: Utilizing GPU matrix acceleration, it transcribes 1 hour of audio in approximately 1 minute on average.
• Global Languages: Supports auto-detection and precise transcription for over 90 languages.
• Strict Data Privacy: All uploaded raw media files are immediately and automatically destroyed after processing, and are never used for model training.

Welcome to ScribeToAny, where turning voice into text is easier than ever!