Audio transcription for hundreds of formats – not just MP3, WAV, or M4A

Automatic audio transcription for demanding tasks

Looking for powerful audio transcription? You’ve come to the right place! RiverScript is an AI app for automatic audio and video transcription, designed specifically for complex tasks. RiverScript supports three types of transcription:

• Transcription of audio and video files up to 50 GB and 8 hours long,
• Transcription of your own voice from a microphone,
• Transcription of audio from Live Recordings from any app.

RiverScript lets you transcribe files in literally hundreds of formats, including quickly and reliably converting even large, high-quality WAV files to text, transcribing M4A audio from Apple, and processing multiple files at once with batch/bulk transcription. RiverScript also has a lightning-fast desktop app for Mac and Windows that compresses and makes files several times smaller before uploading them to the server, making it 3–5 times faster on average than any browser-based transcriber.

What file formats can RiverScript transcribe?

Can it convert MP3 to text?

To decode files, our platform works directly with the open-source FFmpeg library, so we support several hundred audio and video formats and codecs. We have tested the most popular transcription formats, so below you can find a list of audio and video formats that RiverScript is guaranteed to transcribe.

Audio file formats that RiverScript is guaranteed to support for transcription

• MP3, WAV, FLAC, AAC, OGG, WMA, M4A, Opus, AIFF, ALAC, APE, M4B, WavPack, AC3, AMR, and more

There are also hundreds of other formats and codecs that our app can decode and work with – you can find the full list in the FFmpeg documentation.

Please note that if you use the web version of the app in a browser rather than the desktop client, your browser may sometimes prevent you from uploading a file in an uncommon format because it does not recognize it as an audio file. The solution is simple – just transcribe it directly in the desktop app, and you’ll bypass your browser’s limitation.

How does RiverScript turn audio to a transcript?

RiverScript synchronizes the best modern AI transcription models on the market. To convert audio to text, we use models from ElevenLabs, OpenAI, and Deepgram. We don’t just use ready-made models – we synchronize how they work together and apply double filtering. First, we filter noise from your audio and prepare it for the model so it can recognize speech more accurately. Once the transcript is returned, we filter it as well to ensure verbatim text quality. We also use a specialized text model to split the text into paragraphs and add punctuation.

What else can it do?

RiverScript also offers transcription with timestamps for creating subtitles, as well as speaker diarization (available only with the premium model). RiverScript has many additional features, such as translating transcripts into 100 languages using an AI model from OpenAI, chatting with your transcript, and summarization. RiverScript uses a freemium model, so you can transcribe audio for free immediately after signing in to the app.

Who can benefit from audio transcription?

Transcribing audio online is a modern way to save time and effort. It is used by entrepreneurs, content creators, students, journalists, lawyers, and many others. With RiverScript transcription, you can:

• transcribe long files up to 8 hours,
• transcribe large, high-quality 4K videos up to 50 GB,
• transcribe lectures and webinars,
• create verbatim transcriptions of marketing interviews, journalistic interviews, academic research, and focus groups,
• transcribe religious sermons, voice messages, and voice notes,
• create AI transcriptions with timestamps for YouTube videos,
• and much more

RiverScript has a user-friendly interface and lets you transcribe any audio in literally one click.