Transcribe YouTube Videos on iPhone
Search for "transcribe YouTube video" and you'll find YouTube's own auto-captions: free, in about 70 languages, decent for clean English — a fine answer for reading along in the browser.
But the captions are bound to YouTube. They can't be exported as a document, there's no summary, there's no translation into your own notes, and they don't exist for every video: complex audio, unsupported languages, and overlapping speakers are dropped. Auto-translated subtitles degrade fast — a 2023 University of Rochester study found comprehension-affecting errors in roughly 23% of segments for Romance languages and more than 40% for Japanese and Korean — and 2026 vendor estimates put clear-English captions at 90–95% accuracy, falling to 60–75% with background music or noise.
Audio Notes AI targets the file, not the website: if you can get the video into Files, you can transcribe it. Save the video to Files from the YouTube app or site, or screen-record playback for videos that can't be downloaded, then import it. Whisper transcribes server-side, and you get a transcript you own, plus a summary and translation. There is no built-in YouTube downloader and no URL box — the honest scope of transcribing video to text on iPhone is the file you already have.
How to Transcribe a YouTube Video on iPhone
The workflow has three parts: save the file, import it, study the transcript. All three stay within iOS and the app.
1. Save the video to Files
In the YouTube app, open the video, tap Share, and choose Save to Files. On the YouTube website in Safari, the same share/download path puts the file in Files. For videos that can't be downloaded — copyright-restricted content, some Shorts — screen-record playback instead: open Control Center, start a screen recording with the microphone off, play the video with sound, then stop. The clip lands in Photos, ready to export to Files. Note that App Store policy restricts third-party apps from downloading YouTube audio directly, and Audio Notes AI doesn't try: the app's own import guide describes the same save-to-Files workflow.
2. Import the file and Whisper transcribes it
Choose the video in the Files picker inside Audio Notes AI, or hand it over through the iOS share sheet. The app extracts the audio track on-device to M4A and uploads it; openai/whisper-large-v3-turbo transcribes it server-side, with large-v3 as fallback, auto-detecting the language. The picker has 100 entries — 99 Whisper languages plus Cantonese — if you'd rather set it explicitly, which helps with mixed-language videos. Transcription needs internet, the app is iOS-only, and files cap at 90 minutes per recording; a longer documentary splits into two imports.
3. Read, summarize, translate, and keep the transcript
The transcript stays in the source language and is searchable and copyable. When the detected language differs from your preferred language, it is automatically translated and summarized into your preferred language in the same pass — no toggle to find. Export the result as a PDF, or share it through a link encrypted with AES-256-GCM on your device: the key rides in the URL hash and links expire after 24 hours. For lecture material, the same file flows into the course video and lecture video workflows.
Honest comparison: for clean single-speaker English, Whisper and YouTube's captions land in the same accuracy neighborhood, and we won't pretend otherwise. The difference is what you can do with the text — search it, summarize it, translate it, keep it after the video is gone — and the file-first mechanics in the video-to-text guide apply to TikTok clips and MP4 files alike.
| Method | What you get | Limits | Best for |
|---|---|---|---|
| YouTube auto-captions | Free transcript in ~70 languages | Only in the video's default language; no export, summary, or translation; not available for every video; quality varies | Reading along in the browser |
| Auto-translated subtitles | Translated captions over the video | Comprehension errors in ~23% of segments (Romance) and 40%+ (Japanese/Korean) per a 2023 study; still stuck in YouTube | Casual watching in another language |
| Transcript-grabber websites | Caption text pasted onto a page | Caption text only — no summary, no translation, no file of your own; the video must stay on YouTube | Quick copy of a caption line |
| Audio Notes AI | Whisper transcript, summary, translation, PDF export, encrypted share links | Requires the file in Files (save or screen-record); 90-minute cap; no speaker attribution | A transcript you own, search, and keep |
| Manual notes | Your own paraphrase | Slow; misses detail; errors compound over a long video | Short clips where gist is enough |
FAQ
Are YouTube auto-captions free?
Yes, when they exist. YouTube generates auto-captions for supported videos in about 70 languages, and it warns that they can misrepresent speech due to mispronunciations, accents, dialects, or background noise. They also aren't generated for every video: complex audio, unsupported languages, very long videos, poor sound, and overlapping speakers all drop them. And whatever exists stays inside YouTube.
Can I download YouTube videos to Files?
Yes, when the platform allows it. The YouTube app and site both offer a save/share path to Files, and that's the sanctioned route: App Store policy restricts third-party apps from downloading YouTube audio or video directly, and Audio Notes AI has no downloader. You save the file; the app transcribes what you saved. That division of labor is why the import guide inside the app describes the same workflow.
What if a video can't be downloaded?
Screen-record playback. Start a screen recording from Control Center with the microphone off, play the video with sound, and stop when it ends. The clip saves to Photos; export it to Files and import it like any other file. This is the documented workaround for videos with no download option, and it's the same capture path used for short-form clips in our TikTok transcription guide.
Why is my transcript better than YouTube's captions?
For clean single-speaker English, the two are comparable — be skeptical of anyone claiming a huge accuracy gap there. The real differences: Whisper covers 60+ languages the captions skip, the transcript doesn't depend on YouTube's generation rules, and you get a summary, translation into your preferred language, PDF export, and encrypted sharing. Captions are a viewing feature; this is a document. Our speech-to-text app comparison covers where Whisper-based tools sit honestly.
Can I transcribe a YouTube lecture for study notes?
Yes, and it's one of the most common uses. Save or screen-record the lecture, import it, and you get a searchable transcript with a summary — the raw material for study notes on the phone you study from. The same mechanics power online course notes, and any-language video follows our any-language guide. Audio-only clips, like narrated shorts, import through the voice-message workflow instead.
Audio Notes AI imports video files — mp4, mov, m4v, avi, and mkv — extracts the audio, and transcribes it server-side with Whisper (large-v3-turbo, large-v3 fallback) in 60+ languages. Every account includes 60 minutes of lifetime transcription for free; Pro ($5.99/month) unlocks 10,000 minutes per month with a 90-minute cap per recording. Transcripts are translated and summarized automatically into your preferred language, and shared notes are encrypted with AES-256 on your iPhone before upload.
More Video Guides
- Transcribe Video to Text on iPhone: the hub
- Transcribe Zoom Recordings on iPhone
- Transcribe YouTube Videos on iPhone
- Transcribe MP4 Audio to Text on iPhone
- Transcribe Lecture Videos to Text on iPhone
- Transcribe Video Interviews on iPhone
- Transcribe Google Meet Recordings on iPhone
- Transcribe Teams Meetings on iPhone
- Transcribe Podcast Episodes to Text on iPhone
- Transcribe Online Course Videos on iPhone
- Transcribe Webinar Recordings on iPhone
- Transcribe MOV Files to Text on iPhone
- Transcribe TikTok Videos to Text on iPhone
- Transcribe Audio in Any Language (60+ languages)
- Transcribe Voice Messages on iPhone: all 10 apps compared
Download Audio Notes AI free on the App Store — 60 minutes of lifetime transcription included.
Transcribe & Secure Your Audio
Record voice notes, generate instant summaries, and share them via secure expiring links with zero-knowledge encryption.
Download on App Store