Audio to Text
Transcribe interviews, podcasts, lectures and voice memos in your browser. Edit the transcript and export TXT, SRT or VTT.
Select the language spoken in the recording; it is independent of the page language.
First use needs internet to download the Whisper speech model from Hugging Face and the AI runtime from third-party jsDelivr CDNs. Your browser may cache these files. Keep the tab open; processing speed depends on your device.
Your audio and transcript stay on your device and are not uploaded.
First use needs internet to download the Whisper speech model from Hugging Face and the AI runtime from third-party jsDelivr CDNs. Your browser may cache these files. Keep the tab open; processing speed depends on your device.
Review the result: accuracy is not guaranteed. Speaker identification is not supported. This tool does not translate, summarize or separate speakers.
Transcribe an audio recording in three steps
- 1
Choose audio
Choose an audio recording with clear speech from your device.
- 2
Transcribe audio
Select the recording’s spoken language and start local transcription.
- 3
Download TXT
Review and edit the text, then copy or download it.
Ways to use your transcript
- Turn an interview recording into searchable text.
- Draft podcast transcripts for editing and accessibility.
- Review lecture recordings or turn voice memos into notes.
Output examples
These examples show the output format. Your result depends on the source file and the settings you select.
Hello, thanks for watching.1
00:00:00,000 --> 00:00:03,000
Hello, thanks for watching.
WEBVTT
00:00:00.000 --> 00:00:03.000
Hello, thanks for watching.
Edit each segment. Subtitle timestamps are preserved.
If processing fails
- The audio could not be decoded. The file may be damaged, encrypted or use an unsupported codec.
- The speech model could not load. Check your connection and try again.
- Review the result: accuracy is not guaranteed. Speaker identification is not supported. This tool does not translate, summarize or separate speakers.
Keep this tab open. Try a shorter recording with a supported codec and close other demanding tabs before trying again.
Before sharing your result
Play the downloaded file and check the transcript. Keep your original recording. Local conversion does not publish a sharing link.
Frequently asked questions
Is my audio recording uploaded?
Your audio and transcript stay on your device and are not uploaded. First use needs internet to download the Whisper speech model from Hugging Face and the AI runtime from third-party jsDelivr CDNs. Your browser may cache these files. Keep the tab open; processing speed depends on your device.
Which audio recordings can I transcribe?
One audio file, up to 100 MiB and 5 minutes. Supported formats: MP3, M4A, WAV, AAC, OGG, OGA, Opus, FLAC, WebM, WebA.
Which language should I choose?
Select the language spoken in the recording; it is independent of the page language.
How accurate is the transcript?
Review the result: accuracy is not guaranteed. Speaker identification is not supported. This tool does not translate, summarize or separate speakers.