Extract SRT Subtitles from Video Locally

Extract SRT subtitles when a video contains a selectable text caption track and the next task needs that timed text outside the container. This page does not listen to speech and create a transcript. Instead, browser-based FFmpeg selects one existing subtitle stream by its subtitle-track index and writes a separate SRT file locally. Index 0 means the first subtitle stream, index 1 means the next subtitle stream, and the numbering does not include video or audio tracks. The distinction matters in a multitrack MKV or MP4: a player can label streams by language while the underlying order remains different. Start with the first subtitle track, inspect the downloaded text, and retain the source media because the export is a working caption derivative rather than a complete container record.

1. 1

   Identify a selectable text track before entering an index

   Play the source in an application that exposes a subtitle menu and note how many caption choices it offers. Choose index 0 first when you do not know their internal order. If the downloaded SRT contains the wrong language or another editorial version, return to the page and try index 1, then the next number. The index counts subtitle streams only, so a file with one video stream, two audio streams, and two captions still uses 0 and 1 here for the captions. This simple control is deliberate: it avoids guessing a language label from a file name. It does not inspect the container or show a full stream report, so use a dedicated media inspector when accurate program identification is critical.

2. 2

   Convert supported text cues to the simpler SRT structure

   The selected subtitle stream is written as SRT, a widely recognized text format made of numbered cues, timestamp ranges, and dialogue lines. It is well suited to editing, translation, review, and many caption-upload workflows. The conversion preserves the timed text that a compatible decoder can read, but SRT is not a full styling container. ASS fonts, positions, drawing commands, karaoke effects, and detailed layout are not represented as equivalent SRT features. WebVTT settings and other rich annotations may also simplify. Bitmap-based subtitles such as Blu-ray PGS are images rather than directly editable text; turning those into SRT requires OCR, language review, and timing quality control, which this focused browser tool does not claim to perform.

3. 3

   Review the SRT as editorial data, not just a downloaded file

   Keep the tab open while the local job finishes, then download the SRT and open it in a plain-text editor. Check the first cue, a middle section with fast dialogue, and the final cue against the source video. Pay particular attention to accented characters, Chinese, Japanese, Korean, right-to-left text, music labels, speaker prefixes, and cues split across a cut. If the export came from a styled track, decide which formatting must be restored in the destination workflow. Name the file with language and purpose, for example project-en-review.srt, and keep it next to the approved source rather than assuming the exported text is a replacement for the original container, styles, metadata, or accessibility approval record.

Embedded subtitle streams are excellent for playback, but they are awkward when a translator, editor, accessibility reviewer, or publishing platform needs a standalone text file. A separate SRT can be opened in a text editor, sent for terminology review, attached to a caption-upload form, compared with a new edit, or kept beside the video as an editorial asset. Extracting an existing stream also avoids inventing a transcript from audio when timed captions already exist. The stream index makes the choice explicit in files with several language tracks, and the local workflow keeps the media file out of a hosted conversion queue during the export.

The usefulness of SRT comes from its simplicity, which is also its limit. It describes timing and plain text well, but it cannot carry every feature that a source subtitle format or player may use. A visually rich ASS track can look correct in a desktop player while its SRT export intentionally lacks font, color, position, and effect instructions. Likewise, a bitmap subtitle may look like text to a viewer but does not contain editable characters. This page is therefore best treated as a controlled conversion of one known text subtitle stream. Inspect the result in the target workflow and retain the original subtitle stream or source file whenever styling, legal captions, or delivery history matters.

The subtitle-track index is the only content-selection setting. Enter 0 for the first subtitle stream, 1 for the second, and continue only after verifying the prior result. The page maps that one subtitle stream and writes an SRT file; it does not copy video or audio into the download. It also clears container-level metadata from the text export so the result stays a focused caption file. There is no language dropdown because the source stream order, not a guessed label, determines the selection. There is no styling toggle because SRT cannot faithfully express all advanced ASS or WebVTT formatting. Use a media inspector, subtitle editor, OCR system, or authoring tool when your job needs those capabilities.

First verify that the extracted file is the intended language and editorial version. Then compare cue timing with the source at the start, around cuts, and at the final cue. Open the SRT in the application that will receive it to check character encoding, line breaks, and any maximum-line or reading-speed rules. When the source used styling, verify that the plain-text result still meets the next workflow's needs. Preserve the original media and source captions because an SRT export may intentionally omit style, container metadata, alternate tracks, or accessibility flags.

• Use 0 for the first subtitle stream, then increase the index one step at a time after checking the resulting text and timing against the source video.
• SRT carries numbered cues, time ranges, and plain dialogue text. ASS styling, positioning, font choices, and advanced WebVTT features do not become equivalent SRT styling.
• Open the downloaded SRT in a text editor and in the actual destination player. Check encoding, cue order, line breaks, non-Latin characters, and the first and last timestamps.

The typical path uses browser-local FFmpeg: the selected video stays in the local browser workspace while the chosen caption stream is written into an SRT download. No usual account or remote conversion queue is needed. That can be useful for interview footage, internal meetings, training content, client demonstrations, localization reviews, and unreleased media where both audio and its written dialogue are sensitive. Keep the browser tab open until the job finishes and handle the new text file with the same care as the source, because plain subtitles are easy to search, copy, translate, and forward.

A separate SRT can make personal or confidential content easier to discover than a video alone. Names, contact information, health details, legal statements, unpublished product terms, or copyrighted dialogue may be visible in a few kilobytes of text. Local processing does not remove obligations around consent, access control, retention, accessibility, or licensing. Review the exported cues, name the file accurately, keep it in the approved location, and share it only with people who are meant to receive the spoken content in searchable form. Use an audited workflow when the source belongs to a regulated or contractual process.

Move to subtitle-authoring, media-inspection, OCR, or quality-control software when you need a stream list with language and disposition labels, automatic track selection, bitmap-subtitle recognition, ASS style preservation, multiple exports, batch processing, translation memory, speaker identification, reading-speed checks, forced-caption handling, or a formal accessibility report. Use a transcription system when no usable subtitle stream exists and text must be created from audio. Very large media may also be uncomfortable in a browser even though the output SRT is small, because the source must still be read locally. This tool is for one known compatible text stream, one explicit index, one SRT output, and a manual review in the real destination workflow.

• The desired captions are PGS, DVD, DVB, or another bitmap subtitle format that needs OCR rather than a direct timed-text conversion.
• You need automatic track discovery, language labels, forced-caption interpretation, styling preservation, translation, transcription, or a verified multi-track archive.