WaperVideo to text and podcast transcripts
Paste a link and get a proofread transcript where every line plays back from its moment, with a summary and chapters. Save it to your knowledge base and keep going with your agent.
Your first audio or video is free, up to 60 minutes.
See it work
A transcript you can trust
Every word stays, in the order it was said. Waper fixes typos, names and punctuation; it doesn't summarize or rewrite.
Captions first, then proofread
If the video has captions, Waper starts from them; if not, it transcribes the audio. Then it proofreads the text without changing what was said or its order. Any paragraph it corrected can still show the original wording.
Every line plays back
Each paragraph has a timestamp. Select one and the player jumps there and plays. On YouTube and audio links, the line being played is highlighted as it goes.
Search it, read it in your language
Search inside the transcript. If it's in another language, read the translation, or both, with each line over its translation.
Paste the link you already have
Any of these works, or upload a file in Notetaker. You see the length and price before anything starts.
YouTube to text
Videos, Shorts and youtu.be links. Waper uses the video's captions when it has them and transcribes it when it doesn't. The video plays on the page, next to its transcript.
Bilibili videos to text
bilibili.com/video/BV… and b23.tv links. For a video in parts, only the part in your link is transcribed. The creator's subtitles or Bilibili's AI subtitles are used when there are any. Paid content, such as member-only videos and anime, isn't supported.
Podcast transcripts
Paste a podcast episode link or a direct link to an audio file, like an mp3 or m4a. Podcasts rarely have captions, so the whole episode is transcribed, then proofread. Play it back on the page.
Recordings and video files
Upload audio or video from your computer in Notetaker. The file is deleted once it's transcribed, and only you can see the transcript. Since the file isn't kept there's no player: timestamps take you to the right place in the transcript.
Many other video sites work too. Paste the link: if Waper can't read it, you're told before anything is charged.
When the transcript is ready
Summary and chapters
An overview, key points and chapters, each with a timestamp. Select one to jump to that moment.
Ask your agent
Ask about it and the answer cites the moments it used. Have it list the action items, or write it up as an article saved to your knowledge base.
Save to your knowledge base
One click makes a note with the source, overview, key points and chapters. Select lines in the transcript to add them as quotes, each linked to its moment.
Export
Download the transcript as Markdown, SRT subtitles or plain text.
Kept in your library
Each one is an item in your library, so your agents can find it and cite it when they answer later.
Automatic in Radar
When a brief from one of your watches includes a podcast or video, Waper transcribes it in the background and rewrites that entry from the transcript, at no cost to you.
About Radar
Pricing
6 credits a minute, with the transcript, proofreading, summary and translation included.
See plans and pricing- Your first audio or video is free, up to 60 minutes.
- Anything that costs credits shows its length and price first, and starts only when you select Start.
- If someone already transcribed the same public link, you get it right away, free.
- If it doesn't work, nothing is charged.
- Up to 60 minutes each on the free plan, 600 minutes on a paid plan.
Questions
- Which sites and files does it work with?
- YouTube, Bilibili, podcast episodes and direct audio links, plus audio and video files you upload. Many other video sites work too; if a link can't be read, you're told before anything is charged. Live streams and playlists aren't supported.
- How long can it be?
- Up to 60 minutes each on the free plan and 600 minutes on a paid plan. If it's longer, you're told before anything starts.
- How long does it take?
- Videos with captions are usually done in seconds. Ones that need transcribing take longer: a 30-minute video is usually ready in about a minute. You don't have to wait on the page; Waper tells you in your chat when it's done.
- How accurate is it?
- Waper starts from the platform's captions when there are any and transcribes the audio when there aren't, then proofreads typos, names and punctuation without changing what was said or its order. A paragraph it corrected can still show the original wording. It doesn't tell speakers apart yet.
- How much does it cost?
- 6 credits a minute, counted by the started minute, with the summary and translation included. Your first audio or video is free up to 60 minutes, a public link someone already transcribed is free, and nothing is charged if it fails.Plans and pricing
- What happens to files I upload, and who processes them?
- Uploaded files are deleted within 12 hours, once they're transcribed, and only you can see their transcripts. Speech is transcribed on Waper's own servers or, when you're waiting for the result, by ElevenLabs, Alibaba Cloud (through ZenMux) or Google Gemini (through OpenRouter), then proofread by a language model. Your content isn't used to train models. Transcripts of public links are shared by everyone who transcribes the same link.Privacy policy
- How does it fit with my library and knowledge base?
- Everything you transcribe is an item in your library, and Notetaker is where you work with it. Save to knowledge base turns one into a note you can edit. Your agents can search transcripts and cite them in answers.
How to use Notetaker
Adding links and files, playing back, notes and export, step by step.
