UtiliTools Pro

Free online tools

🌐 English ▾
☰
Whisper AI · local processing

AI audio-to-text transcription

Convert speech to text with Whisper directly in your browser and download the transcript as TXT or SRT subtitles.

Select an audio file

MP3, WAV, M4A, OGG, WebM, or another format your browser can decode. Maximum 100 MB and 30 minutes.

Privacy:the audio is decoded and transcribed on your device. The first run downloads the Whisper model from the internet, but your file is not sent to UtiliTools Pro.

Convert voice to text with Whisper

The tool decodes audio in your browser, converts it to Whisper’s sample rate, and runs speech recognition locally. Review or copy the text, and download timestamped subtitles.

Whisper in browser

Speech recognition with a multilingual model running locally.

Language detection

Leave language detection on automatic or choose a language manually.

Private audio

The file does not need to be sent to UtiliTools Pro servers.

Editable text

Check and modify the transcription before downloading.

SRT subtitles

Generate subtitle blocks from timestamps.

Copy or download

Copy the result or save it as TXT and SRT.

How to transcribe audio

  1. 1

    Select an audio file compatible with your browser.

  2. 2

    Leave language detection on automatic or select the spoken language.

  3. 3

    Click “Transcribe with AI” and wait for Whisper to process the audio.

  4. 4

    Check the text and copy it or download it as TXT or SRT.

Without sending your audio to UtiliTools Pro

Whisper runs in your browser through Transformers.js. The model downloads on the first run, but audio is processed locally without a paid transcription API.

Performance and accuracy

Whisper Tiny balances speed and memory use. Clear audio with little background noise usually gives better results. Longer tasks may take several minutes, depending on your device.

Frequently asked questions

Does transcription use Whisper?

Yes. It uses an ONNX version of Whisper Tiny compatible with Transformers.js for speech recognition in your browser.

Is the audio uploaded to a server?

It is not uploaded to UtiliTools Pro servers. Your browser downloads the model and processes the audio locally on your device.

What languages does it recognize?

Whisper supports multiple languages. Use automatic detection or select a common language such as Spanish, English, French, German, Portuguese, Dutch, Arabic, Korean, Japanese, or Chinese.

What audio formats can I use?

Support depends on your browser, but MP3, WAV, M4A/AAC, OGG, and WebM usually work. The browser’s audio features must be able to decode the file.

Can I download SRT subtitles?

Yes. The tool requests timestamps from Whisper and generates an SRT file. If the model does not return segments, it creates a single block spanning the full duration.

Does the first transcription take longer?

Yes. The first time you have to download and prepare the Whisper model. The following transcripts can take advantage of the files stored in cache.

Related tools

AI text to speech

Convert text to speech using neural voices and download it as WAV.

Extract audio from video

Extract audio from a video file before transcribing it.

Cut audio

Select only the fragment you need to transcribe.

MP3 to WAV

Convert audio to WAV directly in your browser.

Change volume

Adjust the audio level before processing.