Novus Stream Solutions
All tutorials

Novus Convert

Convert captions between SRT, VTT, SBV, and LRC

Four caption formats, four different jobs. SRT and VTT for video, SBV for YouTube, LRC for synced lyrics — all converting locally in the browser.

Beginner6 minOpen the toolDocumentation
SRT, VTT, SBV, and LRC caption files converting between one another locally in the browserPin it

Caption formats are a small corner of file conversion with a disproportionate ability to waste an afternoon. They all express the same idea — text with timestamps — in incompatible syntax, and the platform you are uploading to invariably wants the one you do not have.

Novus Convert handles four of them locally: SRT, VTT, SBV, and LRC. SBV and LRC joined in the 24 July 2026 release, and all four convert to and from each other in browser memory.

Contents
  1. 1.1. Know which format goes where
  2. 2.2. Convert locally, not through an upload
  3. 3.3. Expect styling to be lossy in one direction
  4. 4.4. Watch the timestamp conventions
  5. 5.5. Use LRC when the timing is per-line, not per-scene
  6. 6.6. Batch a whole series at once

Two ways to finish

Video captions

SRT and VTT, plus SBV for YouTube.

Synced lyrics

LRC, for players that show words as they are sung.

  1. 1

    1. Know which format goes where

    SRT is the lowest common denominator — nearly everything reads it. VTT is the web standard, required by HTML5 video and carrying styling and positioning that SRT cannot express. SBV is YouTube's own format. LRC is the lyrics format, designed for music players that highlight words as they play.

    Picking correctly saves the round trip. Uploading an SRT where VTT is required usually fails silently rather than loudly, which is the worst kind of failure.

    • SRT: universal, simplest, no styling.
    • VTT: the web standard, supports styling and positioning.
    • SBV: YouTube's format.
    • LRC: synced lyrics for music players.
  2. 2

    2. Convert locally, not through an upload

    Caption files are small, which is exactly why so many online converters treat them casually — you upload the file, it processes on a server, you download it back. Novus Convert does it in browser memory instead, so the file never crosses the network.

    This matters more than the file size suggests. Captions for unreleased material, internal training video, or a client's content are all things you would rather not hand to a free web service.

  4. 3

    3. Expect styling to be lossy in one direction

    VTT can express positioning, alignment, and cue styling. SRT cannot. Converting VTT to SRT therefore drops that information, and no converter can preserve what the target format has no syntax for.

    Plan the direction of travel accordingly: author in the richest format you need and convert down, rather than converting up and expecting styling to reappear.

  5. 4

    4. Watch the timestamp conventions

    The formats disagree about separators and precision — commas versus periods for fractional seconds, different bracket conventions, different tolerance for overlapping cues. These are exactly the details that make hand-editing a caption file error-prone and a converter worth using.

    After converting, spot-check the first cue, the last cue, and any cue you know sits at an awkward boundary. Those three catch nearly everything.

  6. 5

    5. Use LRC when the timing is per-line, not per-scene

    LRC exists because lyrics need denser, line-level timing than a subtitle track usually carries. If you are building anything that displays words in time with music, LRC is the format designed for it and converting to it from SRT is the fast path.

    It pairs naturally with the lyric video tooling in Novus Visualizers at visualizers.novusstreamsolutions.com, which does word-synced captioning on-device.

  7. 6

    6. Batch a whole series at once

    Caption work is rarely one file. The mixed batch queue takes a folder of episodes, converts each, and validates each result before enabling its download — and because outputs are chosen per row, a batch can send some files to VTT and others to SBV in one pass.

    The full route list lives in the conversion directory at /conversions, which is generated from the live capability manifest rather than maintained by hand.

Author rich, convert down

VTT carries styling that SRT cannot express, so convert downward rather than expecting styling to survive a round trip. Spot-check the first cue, the last cue, and any awkward boundary after converting. Use LRC when timing is per-line rather than per-scene, and batch a series in one pass with per-row outputs.

Try Novus ConvertRead the docs

Frequently asked questions

Quick answers to common questions about this topic.

Which caption formats are supported?

SRT, VTT, SBV (YouTube), and LRC (synced lyrics), converting to and from each other. SBV and LRC joined in the 24 July 2026 release.

Will styling survive a VTT to SRT conversion?

No. SRT has no syntax for positioning or cue styling, so that information is dropped. Author in the richest format you need and convert downward.

Are my caption files uploaded?

No. Conversion runs in browser memory. The file never crosses the network, which matters for unreleased, internal, or client material.

What is LRC for?

Synced lyrics — line-level timing for music players that highlight words as they play. It is denser than a subtitle track and is the right target if you are building anything that displays words in time with music.