Subtitle Generator

BETA

Transcribe audio or video, refine every cue, and export timed subtitles without uploading your media.

Quick answer and specifications

The core workflow runs in this browser. Check the supported formats, published limits, and any on-demand model download before you start.

Inputs
WAV, MP3, M4A, AAC, OGG, Opus, FLAC, MP4, MOV, M4V, & WebM
Outputs
SRT, VTT, txt, & JSON
Processing
On this device, in the browser
Model languages
100 languages — see the language selector
Per-file limit
200 MB
Files per run
1
Media duration limit
60 min
On-demand model download
99 MB
Last verified

The interface language does not determine the model language. Recognition and pronunciation quality vary; check the result.

Your media and transcript stay on this device. Speech models are downloaded to your browser and cached for later use.

Drop one audio or video file

Up to 200 MB and 60 minutes. WAV, MP3, M4A, AAC, OGG, Opus, FLAC, MP4, MOV, M4V, or WebM.

Select files

Continue with a focused workflow

Video to SRT Subtitle Generator →

How to generate and edit subtitles locally

Aiviko's Subtitle Generator transcribes an audio or video file up to 200 MB and 60 minutes (WAV, MP3, M4A, AAC, OGG, Opus, FLAC, MP4, MOV, M4V, or WebM) into editable, time-coded subtitles in your browser. It combines local audio extraction (Web Audio or ffmpeg.wasm), Silero VAD (~1.8 MB), and a 99-language Whisper tiny model (~104 MB) to build up to 10,000 cues and export SRT, VTT, TXT, or JSON without uploading your media.

  1. 1. Drop one audio or video file (up to 200 MB / 60 min)

    Select a supported audio or video file and pick the spoken language (or Auto detect) and maximum line length.

  2. 2. Run local VAD and Whisper transcription

    The browser extracts 16 kHz audio, detects speech segments with Silero VAD, and transcribes timed cues in isolated workers.

  3. 3. Edit cues and export SRT, VTT, TXT, or JSON

    Preview cues alongside your media, split or merge lines, shift timestamps globally, resolve any timing warnings, and download your subtitle file.

Supported media formats and subtitle workflows

  • Video to SRT/VTT subtitle generation for MP4, MOV, M4V, and WebM files plus WAV, MP3, M4A, AAC, OGG, Opus, and FLAC audio up to 200 MB and 60 minutes
  • On-device voice activity detection (Silero VAD, ~1.8 MB) paired with 99-language Whisper tiny (~104 MB) speech recognition
  • Interactive cue editor with live media preview, global timestamp shifting, cue splitting/merging, configurable characters per line, and overlap validation (up to 10,000 cues)
  • Clean export to SubRip (.srt), WebVTT (.vtt), plain transcript (.txt), and structured JSON (.json)

Runtime budgets and transcription limits

  • Files over 200 MB, media longer than 60 minutes, decoded PCM over 128 MB, or transcripts exceeding 10,000 cues are rejected to protect browser memory.
  • Clips up to 5 minutes decode directly via Web Audio when supported; longer or complex containers load ffmpeg.wasm in a worker and require keeping the tab open.
  • Automated Whisper tiny transcription and VAD segmentation are a draft—always review wording, speaker changes, and cue timing against playback before publishing.

Related speech and audio tools

Speech to Text →

Quickly dictate with your microphone or transcribe short audio clips into plain text.

Text to Speech →

Turn scripts or translated text into spoken WAV audio across 31 supported languages.

Frequently asked questions about Subtitle Generator

Are my video or audio files uploaded to a cloud server?

No. Media validation, audio extraction, voice activity detection, Whisper transcription, and SRT/VTT export all run locally inside your browser tab. Only the versioned WASM/ONNX runtime and model files are downloaded and cached.

Why is exporting disabled when there are cue validation errors?

Exporting is paused if a cue has empty text, a negative start time, or an end time before its start time so you never download a broken SRT or VTT file. Fix the highlighted cue in the editor to unlock export.

What is the maximum video size and duration?

Each run accepts one supported audio or video file up to 200 MB and up to 60 minutes in duration, generating up to 10,000 subtitle cues.

Before you use this tool

Product information reviewed by Aiviko Team

Where processing happens

The selected input is processed in your browser and is not uploaded to an Aiviko processing server.

What may use the network

The page may download code, fonts, runtime files or machine-learning models. Analytics and advertising load only under the choices and conditions described in the privacy and cookie policies.

Results and limits

Speed, accuracy and maximum practical file size depend on the input, browser and device. Review the guide before using the output for an important decision.

Read the full guide and troubleshooting notes →