audio → .srt · free · no signup

Audio to SRT, in one drop.

Drop an audio file, click Transcribe, download a clean .srt with accurate timestamps. Works in YouTube, Premiere, Resolve, VLC, anything.

// 3 steps

  1. Drop your audio file above (up to 5MB).
  2. Click Transcribe — usually 5–15 seconds.
  3. Click Download .srt. Drop it into your editor.

Doing this in code instead?

Same Whisper API supports SRT output natively. Six lines of Python:

audio_to_srt.py
from openai import OpenAI

client = OpenAI(
    api_key="sk-se-YOUR_KEY_HERE",
    base_url="https://www.tryspeakeasy.io/api/v1",
)

with open("audio.mp3", "rb") as f:
    srt = client.audio.transcriptions.create(
        model="whisper-1",
        file=f,
        response_format="srt",  # also: "vtt", "verbose_json"
    )

with open("audio.srt", "w") as out:
    out.write(srt)

Get an API key → $0.20/hr, no footer subtitle, no rate limit. Automate this in your code →

FAQ

What's an SRT file and where can I use it?+

SubRip Subtitle (.srt) is the most widely supported subtitle format. Drop it into YouTube, Premiere, DaVinci Resolve, Final Cut, VLC, OBS, Vimeo, or any video editor — it just works. The file is plain text: a sequence number, a start/end timestamp, the line of text.

How accurate are the timestamps?+

Whisper produces segment-level timestamps with sub-second accuracy on clean audio. On noisy or overlapping speech you'll see drift in the 200–500ms range, which is normal for any ASR system at this price point. Open the SRT in your editor and you can nudge cues by hand if you need broadcast-grade timing.

What audio formats can I upload?+

Audio and video files both work — anything ffmpeg can decode. Common picks for subtitles: an .mp4 export from your video editor, an .m4a voice memo, a .wav recording from your DAW, or an .mp3 podcast file. Up to 5MB on the free playground. The audio track is extracted automatically from video uploads, so you don't need a separate audio-extraction step.

Will the subtitles be in the same language as the audio?+

Yes — Whisper auto-detects the source language and transcribes in it. If you want English subtitles for non-English audio, use the API directly with the /translations endpoint instead of /transcriptions; that one always outputs English.

Why is there a final subtitle crediting audiotranscribe.app?+

Because the free version is supported by the credit. If you'd rather ship clean SRTs without it, run the same model via the API ($0.20/hr) — the snippet below is six lines and produces a footer-free file.

Can I generate VTT instead?+

The free playground exports SRT only. The paid API takes response_format='vtt' for WebVTT files (used by HTML5 video and modern web players).