How to Convert Audio to SRT Subtitles Online

To turn audio into SRT subtitles, open the studio, choose the Speech to Text tab, upload your audio or video file and press Transcribe. You get the full transcript, which you can copy, download as a TXT file, or download as an SRT subtitle file with timestamps.

Use it in your browser; the file is transcribed on our servers, so there's nothing to install and you don't need a graphics card. You sign in with Google, and new accounts get 20 free credits, enough for 20 minutes of audio.

What you get

• The full transcript in the page, with a Copy text button.
• A TXT file with the plain text.
• An SRT file with the same words as subtitles, each with a start and end timestamp, ready to load into a video editor.

How to convert audio to SRT, step by step

1. Open the studio and sign in with Google.
2. Choose the Speech to Text tab.
3. Press Upload a file and choose an audio or video file up to 30 MB and 15 minutes long. The studio shows its length.
4. Under Spoken language, leave Detect automatically or choose the language yourself.
5. Check the cost and your credit balance: 1 credit per minute of audio, rounded up.
6. Press Transcribe and keep the tab open while it works.
7. Read the transcript, then press Copy text, Download TXT or Download SRT.

Files you can upload

• Formats: MP3, WAV, M4A, MP4, MOV, WebM, OGG and FLAC.
• Size: up to 30 MB per file.
• Length: up to 15 minutes per file.

Video files work too. Upload an MP4, MOV or WebM file as it is; you don't need to extract the audio first.

Cut a recording longer than 15 minutes into parts and transcribe each one. The timestamps in each part's SRT start from the beginning of that part, so place each file at the right point on your timeline, or shift its timing in your editor. If a file is under 15 minutes but over 30 MB, export it as compressed audio such as MP3 or M4A, which is usually much smaller than a video or WAV file.

Languages and accuracy

Speech to Text recognizes about 99 languages. It detects the spoken language automatically, or you can set it yourself when you know it. It runs on our own servers with Whisper small, an open-source speech recognition model released under the MIT license.

Accuracy varies with the quality of the recording. Results are usually more accurate with:

• clear speech, recorded close to the microphone;
• little background music, noise or echo;
• one person talking at a time.

Names, technical terms, strong accents and overlapping voices are where mistakes are most likely. The transcript is generated by AI and can contain errors, so read it through, or play your video with the subtitles on, before you publish.

What an SRT file is

SRT is a plain-text subtitle format. Each subtitle in the file has a number, a start time and an end time, and the text shown on screen between those two times. Because it's plain text, you can open an SRT file in any text editor to fix a word. Change only the text lines and leave the numbers and timestamps as they are.

How to add the SRT to a video

Most video editors can import subtitle files: look for an option to import captions or subtitles, choose the SRT, and place it at the start of your timeline so the timestamps line up with the speech. Play it through once and adjust any line that appears too early or too late. Many video platforms also accept an SRT file as captions when you upload a video.

Credits and free minutes

• Cost: 1 credit per minute of audio, rounded up. A clip under a minute costs 1 credit, and a 15-minute file costs 15 credits.
• Free start: you get 20 welcome credits when you first sign in with Google, valid for 365 days. That covers 20 minutes of transcription, spent across files, while each file keeps its 15-minute limit: for example, one 15-minute file and one 5-minute file.
• Shared balance: the same credits pay for Text to Speech and Voice Clone.
• More credits: 1,000 credits cover 1,000 minutes of transcription. Monthly plans and one-time packs are on pricing. Plan credits are valid until the end of each billing month and don't roll over; pack credits never expire.
• Failed jobs: if a transcription fails, its credits go back to your balance automatically.

What happens to your file

Speech to Text runs on our own servers, so your file isn't sent to an outside AI service to be transcribed. Uploaded files and transcripts are kept in our cloud storage for your account for 30 days, so your download links keep working, and are then deleted automatically. After you've made a purchase, new uploads and transcripts are kept for 6 months. Download the TXT or SRT files you want to keep. We don't use your uploads to train models. You can ask us to delete your files sooner by emailing [email protected]. The Privacy Policy has the details.

Questions people ask

Is audio to SRT free here?

You can start for free: the 20 welcome credits cover 20 minutes of audio. After that, transcription costs 1 credit per minute from a plan or a pack, and the cost is shown before you transcribe.

Can I make subtitles from a video file?

Yes. Upload the MP4, MOV or WebM file directly and download the SRT.

How long can the audio be?

Up to 15 minutes and 30 MB per file. Split longer recordings into parts.

Which languages does it support?

About 99. Leave it on automatic detection or pick the language yourself.

Can I edit the subtitles?

Yes. Change the text in a text editor or in your video editor's subtitle tools.

Can I also turn text into speech?

Yes. The Text to Speech tab makes an MP3 from your text; the text to MP3 guide explains how, and the voice cloning guide covers your own voice.