SonaFactEvidence from audio
SonaFactEvidence from audio

Transcription for meetings, sales calls, interviews and podcasts

The same pipeline, different finish line: minutes, a call review, an exact quote, or subtitles and chapters.

MeetingsSales & supportInterviews & researchPodcasts & video
Teams & managers

Minutes without a note-taker

Record the meeting, and get participants, decisions and action items with owners and deadlines — each one linked to the moment it was agreed.

1Upload the recording; the engine tells the speakers apart and you give them names.
2Skim the transcript, fix names, then generate meeting minutes.
3Send DOCX or PDF; anyone can click a timestamp to hear the original.

The minutes arrive with owners and deadlines already filled in — and every one of them carries the second it was agreed, so checking a draft takes a few clicks instead of a second listen.

Meeting minutes · action itemsRemoved 2 items with no source
01:01Anna WhitfieldSend the migration summary to all participantsToday
00:23James CarterConfirm the Munich data hand-over date18 Sep
00:52unassignedBook the follow-up meeting in two weeks—
Meeting minutes
Decisions, participants, summary
Action items
Who · what · due · moment
DOCX / PDF
Shareable with timestamps
Call review
Scored against your criteria
Sales & support leads

Review calls by scorecard

Talk share, pace, interruptions and questions, with a criterion-by-criterion score and the justification behind it. Coaching signals instead of an hour of listening.

1Upload the call and assign roles — AI suggests rep and customer.
2Generate the sales or support report with your scorecard.
3Coach on the three moments that matter, linked in the report.

A scorecard turns an hour of audio into a page you can read — each criterion with a verdict, the reason behind it, and the moment in the call it was based on.

Sales call · BANTScore 72%
04:12BudgetCustomer named a range for the 2027 budgetMet
07:48AuthorityDecision maker mentioned but not confirmedPartial
11:02NeedTwo concrete problems with the current workflowMet
—TimelineNo date discussed for the decisionNot met
Scorecard
BANT, support standard, or yours
Metrics
Talk share, pace, monologue
Objections
With the response given
Next steps
Owner and due date
Journalists, researchers, HR & legal

Quotes you can stand behind

Exact wording with word-level timing, every speaker told apart, and a glossary that teaches the engine your names and terms before it listens — so they come back spelled right.

1Add recurring names to your glossary once; every later recording is transcribed with them in mind.
2Verify low-confidence words, which are underlined for you.
3Export a transcript with timestamps and a participants table, or a file ready for NVivo, MAXQDA or ATLAS.ti.

Every line has its time, so a quote can be played back from the moment it was said, and the words the engine was unsure about are underlined before anyone prints them.

interview-01.mp3 · utterancesEnglish · 2 speakers
12:04ReporterWho made the decision to pause the project?
12:11M. HartleyThe board made it, at the request of the legal team.
12:26M. HartleyWe need to check invoicing in KSeF.
12:39ReporterMhm.Low conf.
Word-level timing
Play from any line
Participants table
Included in exports
Glossary
Your names and jargon, any language
Confidentiality
Encrypted, processed in the EU
Podcasters & video creators

Chapters, subtitles, translations

A chapter list ready for a YouTube description, subtitles that follow readability rules, and translations you can read next to the original before they go out.

1Upload the episode; chapters and a summary come from the transcript.
2Generate subtitles, then add up to four translated languages.
3Jump to anything flagged too fast, shorten it in the transcript, then export SRT or WebVTT.

Chapters land ready to paste into a description, and before you download, the app counts the subtitles running too fast to read and shows where they are.

Chapters · copy for YouTubeEpisode 34 · 58:12
00:00IntroWhat we are covering this week
04:30Chapter 2Why speaker separation changes editing
21:15Chapter 3Subtitle readability, in practice
47:02OutroListener questions and next episode
Chapters
One click to clipboard
Subtitles
SRT and WebVTT
Translations
Up to 4 languages, uncertain sentences marked
Show notes
From the AI summary

Start with one recording you already regret not writing up

Start transcribing for freeSee all features