Audio transcription software for every file on your drive
Client interviews, podcasts, lectures, dictaphone files, recorded calls and video footage all end up as files in a folder. Stenograph accepts them as they are, in the formats your devices produce, and returns transcripts with timestamps and speaker labels, ready to export as documents or subtitles.
Listening
Audio and video formats you can upload
You do not need to convert anything first. Drop the file into the web app or the desktop app for Mac and Windows, and Stenograph reads the audio track directly, whether it came from a phone, a handheld recorder, a conferencing tool or a camera.
- Audio files in MP3, M4A, WAV, AAC, FLAC, OGG and OPUS
- Video files in MP4, MOV, WEBM and MKV
- Phone voice memos and dictaphone recordings
- Exported recordings from video calls
Timestamps on every passage
Each paragraph of the transcript carries the time it was spoken, and you can switch to sentence level timing when you need finer reference points. Clicking a timestamp plays the audio from that second, which makes checking a quote or a disputed phrase a matter of seconds instead of scrubbing through the file. Timestamps stay in the export if you want them, so a colleague reading the Word or PDF version can find the same point in the original recording.
Speaker labels for interviews and group recordings
Stenograph separates the voices in a recording and labels each turn. Rename a label once, for example from Speaker 2 to the name of the person interviewed, and the change applies everywhere in the file. If the same person was split into two labels because they moved away from the microphone, merge the labels in one step. Clear speaker turns make interview transcripts and panel recordings readable without listening again. For researchers and journalists, this also means a quote can be attributed to the right person with confidence, and the timestamp beside it points straight to the evidence in the original file.
- Automatic separation of voices
- Rename once for the whole file
- Merge labels detected twice
- Speaker names included in exports
Long recordings without splitting them
A full day workshop or a three hour interview is one file in Stenograph. You upload it whole and keep working while it is processed; the transcript opens when it is ready. Long recordings are divided into sections by time and topic, so you can navigate a lengthy transcript like a document with a table of contents. The minutes count against your monthly transcription allowance, which is 300 on Solo, 900 on Pro, 3,000 pooled on Team and 10,000 pooled on Enterprise.
SRT subtitles with correct timing
Video creators, trainers and communications teams often need captions more than they need a document. Export any transcript as an SRT file and the subtitle cues are timed to the speech, split into lines of readable length, ready to load into a video editor or a player. Your dictionary applies to subtitles too, so a product name on screen is spelled the way your brand spells it.
- SRT with cue timing from the audio
- Readable line length for captions
- Speaker names optional in cues
Word, PDF and TXT for documents
Beyond subtitles, transcripts export to Word for editing, to PDF for sharing and records, and to plain TXT for any system that needs simple text. You choose whether to include speaker names and timestamps. Teams that also want summaries of their recordings can use the ai meeting note taker, and buyers comparing tools for a whole office will find admin and billing details on the transcription software page.
Getting the best transcript from an audio file
Recognition is strongest on clear audio, and a few habits make a noticeable difference. Record close to the speakers when you can, avoid placing the recorder next to a fan or a keyboard, and add the names and terms you expect to your dictionary before uploading. For recordings in more than one language, leave language detection on so mixed speech is kept as spoken.
- Add expected names and terms before you upload
- Keep the microphone close to the main speaker
- Leave detection on for mixed language recordings
Files are processed, then the audio is deleted
Uploaded audio is processed and then deleted; Stenograph does not keep your recordings. The resulting text is stored only if text history is on, and it can be deleted at any time. Your files are never used to train AI models. Encryption protects every file in transit and every transcript at rest.
Audio transcription questions
Is there a maximum file size?
Large files are accepted, including multi hour recordings and video. If an upload is very large, the desktop app is the most reliable way to send it.
Can I transcribe only part of a file?
Yes. Set a start and end time before processing, and only that range counts against your minutes.
Does the SRT export include speaker names?
It can. Speaker names are optional in subtitle cues, and you choose when you export.
What happens to the video after transcription?
The audio track is processed and deleted, and the file is not stored. Only the transcript text stays, if text history is on.
Which plan suits regular interview transcription?
Pro includes 900 transcription minutes a month for one person. A team of up to five people can share 3,000 minutes on Team.
Upload the file you have been putting off
Choose a plan, drop in your longest recording and export the transcript or subtitles the same day.