Is it true that AI companies are making bank by ingesting, remixing, and repackaging lectures?

At RiverScript, we do absolutely none of that. But we cannot speak for all transcription services, as we don’t know how they process, store, or secure lecture audio and generated transcripts. We always recommend that users thoroughly research a company before uploading their learning materials. RiverScript is strictly a tool designed to capture audio, convert it into text, and help students study. We do not train AI models on university lectures, nor do we sell or repackage data. Here is exactly how we protect your privacy: 1. We do not use your data or share it with anyone to train AI models. Your transcript data belongs solely to you, and only you control it. We comply with the GDPR and operate within the European Union, following EU privacy and data protection standards. All transcripts and account data are stored on our Hetzner servers in Helsinki, Finland, EU. Audio is stored for exactly seven days for your convenience, allowing you to return to it, listen while reading the synchronized transcript, and navigate the audio using timecodes. After seven days, the audio is permanently deleted. It is never shared with anyone after transcription. Audio recorded on your computer using RiverScript Desktop client is also stored on your computer and is not deleted automatically – it remains entirely under your control on your device. 2. We use trusted enterprise API partners under strict data protection agreements. Users’ audio streams are processed solely to generate transcripts and are not retained by these providers. All our subprocessors are transparently listed in our legal documents. 3. RiverScript does not use tracking or marketing cookies. You are completely anonymous to third-party advertisers, and we protect our users from them. 4. Users have complete ownership and control over their data. RiverScript is built strictly for personal education and intellectual growth. We trust our users to respect university intellectual property and use these tools solely to enhance their own learning experience. We also recommend that you carefully review the applicable laws and the lecture organizers’ rules before recording anything. These laws and rules may vary between countries and institutions, and we strongly recommend that you follow them. 5. In accordance with EU law, we allow users to delete any transcript data, permanently delete their account, and export all their transcripts as an archive. All these options are available in the settings.

How does your app recognize chemical formulas and specialized terminology? With other apps, I constantly had to replay lectures and correct nonsense and hallucinations.

We have two hybrid models: Standard Sync Model (OpenAI Whisper v3 + Deepgram Nova-2) and Premium Sync Model (ElevenLabs Scribe v2 + Deepgram Nova-3). Our engineers developed an AI architecture in which models from different providers complement and enhance one another. We also filter both the incoming audio and the text returned by the models in several stages. These solutions were developed entirely by RiverScript and significantly improve the quality of the final transcripts. Both our transcription models have performed exceptionally well when transcribing complex specialized terminology and formulas. However, we recommend Standard Sync Model (Free Start and Pro plans) for most standard needs – including studying. We recommend Premium Sync Model (Premium plan) for the most accurate enterprise-grade transcription when accuracy matters most.

If a professor has an accent, can RiverScript convert their speech into clear text?

Yes. Even RiverScript’s Standard Sync Model (OpenAI Whisper v3 + Deepgram Nova-2) is sufficient for this task. But if the lecturer’s accent is so strong that you can barely understand anything, try Premium Sync Model (ElevenLabs Scribe v2 + Deepgram Nova-3). It delivers exceptionally clean results using the latest technology on the transcription market and converts speech into text while completely eliminating false word matches.

Our professor always bans bots from live lectures in Zoom and Teams, so I can’t transcribe the audio into text for studying. This only makes it harder for many students to learn, especially if you have auditory processing disorder or are studying in a non-native language as an international student. Can the professor kick RiverScript out of the webinar?

No, you don’t need to worry about that. RiverScript Desktop client does not depend on other platforms – it runs independently on your device. It does not integrate itself into your online learning apps, so the professor cannot accidentally ban it. It simply runs on your computer, records the system audio, and then transcribes it. This allows you to study according to your personal learning needs – recording, listening to, and reading lecture transcripts at your own pace and whenever it is convenient for you.

Can RiverScript translate lectures and transcripts? Is it like Google Translate?

RiverScript translates lectures and other transcript text into 100 languages, the full list of which is available in the documentation. RiverScript translates transcripts with AI-level quality – it literally uses a model from OpenAI, so the quality is similar to translating with ChatGPT. But there is no chat length limit, so you can translate even an eight-hour lecture into another language or several languages at once, while maintaining AI-level translation quality. It is much better than Google Translate.

Want to download lectures from Panopto, Blackboard, Kaltura, or Canvas? Record and transcribe them with RiverScript!

A professional AI workflow to grasp the core of a lecture without frying your brain

If you want to get as much as possible out of your lecture material instead of frantically typing every word the professor says, this article is for you. RiverScript is recording and transcription software that can extract audio from online lectures and convert lectures you can’t download into text from any university platform, including Panopto, Blackboard, Kaltura, Canvas, and more. RiverScript Desktop client does this with Live Recording Transcription – it captures your computer’s system audio and uses two specially configured AI Models with speech pre- and post-processing to deliver the highest transcription quality: the Standard model (OpenAI Whisper v3 + Deepgram Nova-2) and the Premium model (ElevenLabs Scribe v2 + Deepgram Nova-3). These AI models recognize professional terminology and formulas without false matches and can even handle a professor with a strong accent. No more brain fry from listening to online webinars every day.

Below is a quick guide to getting started with the professional recorder. Detailed instructions are available for Mac and Windows.

Download Desktop client for Mac or Windows from the download page.
Install it.
Click Start recording in the app.
Start the lecture in your university’s player.
When the class ends, click Stop recording.
Want your lectures as searchable text? Click Transcribe.
Only want to save the recording to your computer? Click Save locally (you can transcribe it later).

If your professor tracks how many minutes you watched the video, the system will still count the lecture as watched even if you were actually drinking coffee while RiverScript recorded the audio stream to create a complete set of text notes.

Study smarter for university exams with RiverScript

Find quotes in lecture transcripts by keyword

You can find specific quotes or terms using keyword search. Click anywhere in the text in the editor, press Ctrl+F on Windows or Command ⌘+F on Mac, and enter a word in the box in the top-right corner. Matching words will be automatically highlighted in the transcript.

Get a deep lecture summary and listen only to what matters

In the AI (Summarize & Ask) tab, choose your language and get a detailed yet concise summary of what the lecture is about. You can also click Key moments and listen only to the parts directly related to the subtopic you’re interested in. All identified subtopics are automatically labeled, so any student can quickly navigate the material without replaying the entire track several times.

You also have access to AI chat, where you can ask it to identify, list, or formulate any information mentioned in the source: terms, formulas, definitions, scientific theories, dates, articles of law, and more.

Study in a foreign language

If you’re pursuing higher education in a language that isn’t your native language or want to broaden your horizons with unique information from foreign-language lectures, translate a transcript converted from audio or video. You can get high-quality translations from 100 languages. The translator is configured to work like ChatGPT, but without its typical drawbacks, such as chat length limits. RiverScript’s translator uses an OpenAI model, doesn’t get lazy with long transcripts, and chooses the right words when translating professional terminology. Tested on medical lectures.

Edit and save your notes for later

You can remove anything that isn’t important to you from the lecture transcript or add your own notes directly in the Editor. (For example, if part of the recording contains an ad and you don’t need it.) You can then save the resulting notes in PDF, DOCX, or TXT format, or as SRT or VTT files. You can also use the Share button to send classmates a link to the text in a group chat, build a personal archive of study notes in Notion, or copy them as Markdown and paste them directly into Obsidian. And we’ll keep adding new integrations to RiverScript.

Capture lectures from Panopto, Blackboard, Kaltura, and Canvas – study for exams more efficiently