Upload
MP3, WAV, M4A, MP4, WebM, OGG, or FLAC. Up to 120 minutes per file.
Turn any recording into a script.
Upload audio or video. Get a timestamped transcript ready for captions, editing, or narration.
A production workflow
Keep the timing when the words become a script.
This example shows the output shape: a transcript with timestamps that can become subtitles or a new narration script.
Sample recording00:18
00:18The carriage held but just ourselves and immortality.
SRTVTTTXTSend to narration
Three steps from recording to script.
The same pipeline behind NarrateHQ's studio, applied in reverse.
Transcribe
A timestamped transcript comes back, viewable as plain text or with timing.
Export or send to narration
Download as SRT, VTT, or TXT, or send the transcript into the studio as a script.
From recording to finished voiceover.
Use the transcript as a starting point, then keep the rest of the production workflow in one place.
RecordingTimestamped transcriptEditable scriptAudio + SRT
See how section replacement works
13 languages. Pick yours for the best result.
Auto-detect works well for the common case, but selecting your language before uploading gives the most accurate result. Auto-detect can occasionally mix up similar-sounding languages.
English
Thai
Spanish
Portuguese
French
German
Japanese
Korean
Chinese (Mandarin)
Vietnamese
Indonesian
Hindi
Arabic
Your audio isn't kept longer than it needs to be.
The uploaded recording and the transcript follow different retention rules.
Source audio deleted on completionThe file you upload is removed when its transcript is ready, whether the job succeeds or fails.
Transcript kept indefinitelyYour transcript stays on your account like any script, until you delete it.
Separate minutesSpeech to Text minutes are measured separately from narration credits.
120-minute file capA hard ceiling per upload, shown before you spend a minute of quota.
Questions before you start.
MP3, WAV, M4A, MP4, WebM, OGG, and FLAC, up to 120 minutes per file.
English, Thai, Spanish, Portuguese, French, German, Japanese, Korean, Mandarin Chinese, Vietnamese, Indonesian, Hindi, and Arabic.
The transcript includes timestamps from the completed transcription job. You can download the timed SRT or send the text into narration.
When a recording has more than one speaker, segments are labeled automatically (Speaker 1, Speaker 2, and so on). Speaker labels are automatically detected and may not always be accurate.
Only upload audio or video that you have the right and consent to process. The supported formats and per-file duration limit are shown before upload.
Transcription uses a different provider with its own cost per minute, so it's metered separately rather than drawn from the same pool as narration.
It's deleted as soon as the transcript is ready. The transcript itself stays on your account until you delete it. See the Privacy Policy.
Yes. Send it to the studio as a script with one click, then generate audio in any voice.