How to Transcribe a Video

Creators and editors use speech-to-text to convert the dialogue inside their video recordings into written documents. Once you convert the audio to text, you can review the content without constantly scrubbing through a video player timeline.

Additionally, transcribing a video is the first required step if you plan to create captions or subtitles to improve the accessibility of your content.

Note that video transcription converts existing spoken audio into text. If you instead need to generate voice narration from a written script, see our guide on text-to-speech for YouTube videos.

Konthora accepts MP4, WebM, and MOV video files (as well as MP3, WAV, M4A, and AAC audio files). When you transcribe audio from a video, Konthora processes the available audio track embedded inside the file.

Before starting, ensure your file fits within the platform limits. The maximum upload size is 100 MB, and the maximum media duration is 10 minutes. Longer videos must be manually divided into shorter files before upload.

For the best audio transcription accuracy, ensure the video's audio track is clear and free from heavy background noise.

Applying timestamps makes it easy to match the written text back to the video timeline. Konthora provides sentence, paragraph, and word-level timestamp modes.

Sentence mode is highly recommended if you are creating subtitle tracks, as it breaks the text into manageable chunks. Paragraph mode is better if you just want to read the transcript like a document. Word-level timestamps provide exact frame-level alignment for precise editing workflows.

Konthora generates a standalone transcript or subtitle file, rather than burning the text directly into the video. You must choose an export format that fits your goal.

If you need a simple readable document, export as TXT. If you want to add captions to video using a player or editing software, you should export as SRT or VTT. Developers and technical users can also export JSON for programmatic access to the data.