NoScribe
Transcribe audio and video files into editable text
In their words
NoScribe is a web app that turns audio and video files into editable text. It accepts MP3, WAV, FLAC, M4A and MP4 files, and the FAQ adds MOV and AVI.
Input and output
- Drag and drop audio or video files, or click to select them
- Choose from more than 100 languages
- Export as .txt, .srt, .vtt, .docx or .pdf
- Exports carry timestamps and speaker labels when applicable
Speakers
Optional speaker diarization labels voices as Speaker A, Speaker B and so on. Expected speakers can be set to auto-detect or to a count from 2 to 10.
Credits and limits
One credit equals one minute of audio or video. The Free plan includes 300 credits and a 100 MB upload limit per file. Pro allows 5 GB per file and Business allows 10 GB.
Accuracy and privacy
The FAQ cites 95-99% accuracy, depending on audio quality and background noise. The site says files are encrypted during upload and processing and are not stored after processing.
Fact file
Generator- Web
- Credits
- 1 credit = 1 minute
- Upload limit
- 100 MB Free, 5 GB Pro, 10 GB Business
- Exports
- .txt, .srt, .vtt, .docx, .pdf
- Accuracy
- 95-99%
What is NoScribe?
Audio and video files go in, and editable text comes out. NoScribe accepts MP3, WAV, FLAC, M4A and MP4 files, and its FAQ adds MOV and AVI. A file can be dropped onto the upload panel or chosen with a click. The panel then shows the file's name, size and format while it uploads, and a language is picked before the Transcribe button starts the job. Exports come as .txt, .srt, .vtt, .docx or .pdf. The .srt and .vtt files are timed caption formats, and each export carries timestamps and speaker labels when they apply.
Speaker labels and language choice
Speaker diarization is off by default and can be switched on. When it is on, the expected speaker count is set to auto-detect or to any number from 2 to 10, and voices are labelled Speaker A, Speaker B and so on. The page notes that separation works best when speakers do not talk over each other. Language support is listed as more than 100 languages. The named examples are English in US, UK and Australian forms, Spanish, French, German, Italian and Portuguese, and the FAQ also names Mandarin, Japanese, Arabic and Hindi.
Credits, upload limits and plans
One credit covers one minute of audio or video. The Free plan includes 300 credits, roughly 5 hours, with a 100 MB limit per file and 5 GB of encrypted cloud storage. Pro allows 5 GB per file and 500 GB of storage, and Business allows 10 GB per file and 1 TB. Credits reset at the start of each billing cycle. The one-time packs carry lifetime validity, and the plan list says Pro and Business exports have no watermark.
Accuracy claims and data handling
The FAQ puts accuracy at 95-99%, depending on audio quality, accents and background noise, and says clear audio with little background noise gives the best results. The site states that files are encrypted during upload and processing and are not kept after processing. Transcripts can be corrected in the built-in editor.