AI Speech to Text - Transcribe Audio with Speaker Labels Online
title
subtitle
audio_upload
audio_upload_tip
diarize_hint
tag_events_hint
keyterms_hint
advanced
public_hint
empty_title
empty_hint
Why Speech to Text?
Turn any recording into searchable text with timestamps and speaker labels
Accurate Transcription
Upload your audio and get a clean, readable transcript with word-level timing for quoting and subtitling.
Speaker Labels
Enable diarization to separate who said what — ideal for interviews, meetings, and multi-host podcasts.
Key Term Boosting
Feed in product names, jargon, and proper nouns so the transcript spells them right the first time.
Ready-Made Exports
Copy the text, download a .txt, or export an .srt subtitle file with properly timed cues.
How to Transcribe Audio
Three simple steps to a finished transcript
Upload Your Audio
Drop in an MP3, WAV, M4A, FLAC, or OGG file up to 60 minutes. Duration is verified automatically.
Pick Language & Options
Set the audio language or let it auto-detect, toggle speaker labels, and add key terms if needed.
Read & Export
Review the transcript by speaker with timestamps, then copy it or download .txt and .srt files.
Use Cases
Built for journalists, podcasters, researchers, and teams that work with recordings
Technical Specifications
Speech to Text converts your audio into a structured transcript with word-level timing, optional speaker labels, and subtitle export.
- Accepts MP3, WAV, M4A, AAC, FLAC, OGG, OPUS, and WebM up to 300 MB
- Maximum audio length: 60 minutes
- Auto-detects language or choose from 19+ languages
- 2 credits per minute — 2.4 with key terms — billed on verified duration
- Export as plain text or timed SRT subtitles
- Transcripts can be used in personal and commercial projects
Frequently Asked Questions
Everything you need to know about Speech to Text
It converts your audio into a readable transcript with word-level timestamps, optional speaker separation, and sound event tags — then lets you copy it or export it as .txt or .srt.
MP3, WAV, M4A, AAC, FLAC, OGG, OPUS, and WebM up to 300 MB and 60 minutes. Language auto-detects by default, or pick from 19+ languages including English, Chinese, Spanish, French, German, Japanese, and Korean.
2 credits per minute of audio, or 2.4 credits per minute when you supply key terms. A 30-second clip costs 1 credit. The exact price is calculated from your file's verified duration and shown before you run.
Key terms are up to 100 words or short phrases — product names, technical jargon, proper nouns — that help the transcript spell them correctly. Each term can be up to five words.
When enabled, the transcript is grouped by speaker so you can see who said what — useful for interviews and meetings. Segments are marked with timestamps like 03:42.
We build standard subtitle cues from the word-level timestamps — breaking lines at speaker changes, 7 seconds, or ~42 characters — so the .srt drops straight into most video editors.
Ready to transcribe?
Upload a recording and get a finished transcript in moments.
Transcribe audio now