TalkScribe

Everything between the recording and the finished document

Six capabilities, one file. Start with the transcript and take it as far as you need — minutes, a call report, subtitles in five languages, or just a clean DOCX.

Upload & processingTranscript editorVoice profilesAI documentsCall analyticsSubtitles & glossaries
Upload & processing

Queue files and watch each stage

Drop a single interview or a week of calls. Each file gets its own language, speaker count, glossaries and toggles, and reports exactly which stage it is in.

Automatic language detection with a confidence figure — “English (98%)”.
Speaker count: auto, an exact number, or a range.
Per-file glossaries, extra names and AI correction.
Plain-language errors with a fix, not a code.
MP3 · WAV · M4A · MP4 · MKV · OGG · FLAC · WMARetry and logs per file
Queue3 files
42:10product-review.mp4Transcribing · 64% · ~2 min leftProcessing
18:24interview-01.mp3Recognising known voices · 88%Processing
07:55support-call.m4aChinese (94%) · 2 speakersDone
old_tape.wmaAudio file not found — locate fileError
Stages: decoding → speakers → language → transcription → word alignment → glossary → utterances → known voices → AI correction.
Transcript editor

Fix a word, rename a speaker everywhere

A waveform coloured by speaker, word-level highlighting that follows playback, and inline editing of everything — text, speaker, and where one utterance ends.

Ctrl+click a word to play from it; Enter to edit, Ctrl+Enter to split.
Low-confidence words are underlined so you know where to look.
Glossary suggestions appear inline: replace with KSeF, or ignore.
Speaker panel with speaking time, share, voice sample and voice menu.
Undo everywhereSearch & replaceFollow playback
product-review · utterances9 utterances · 142 words
00:00Anna WhitfieldGood morning, Ms Weber. Thanks for making time today.OK
00:10Speaker 2I do have a few questions about the schedule.Edited
00:27Speaker 2We need to check invoicing in KSeF.Suggestion
00:44Anna WhitfieldYeah.Low conf.
Short interjections like “yeah” or “ok” are the most common thing to correct.
Voice profiles

People it has met before get their names back

Save a speaker as a voice profile and TalkScribe names them in the next recording. When the audio is poor it becomes cautious and asks instead of guessing.

Recognition badge with a percentage — confirm or reject in one click.
Sensitivity: careful, recommended, or high.
Profiles store a legal basis or consent and a note.
Voice prints are encrypted and never appear in any export.
Biometric dataSwitch off at any timeDelete profile deletes samples
Voice profiles4 profiles
3Anna WhitfieldLast recognised: today · consent on fileActive
2James CarterLast recognised: 12 Sep · consent on fileActive
1Maria WeberSuggested in product-review.mp4 (74%)Suggested
0New from file…Name, consent (required), audio fileAdd
Samples can be deleted individually; “delete all” requires typing to confirm.
AI documents

Minutes, action items, chapters — with receipts

Generate a document from a built-in or custom template. Every item carries a clickable timestamp, and anything the transcript does not support is removed before you see it.

Meeting minutes, action items, summary, chapters, custom templates.
Verification line: “Removed 2 items not supported by the recording.”
Documents mark themselves outdated when the transcript changes.
Output language independent of the recording language.
DOCX · PDF · Markdown · CopyEdit as markdownRegenerate
Action itemsMeeting minutes · up to date
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
🛡 Removed 2 items not supported by the recording.
Call analytics

Scorecards for sales and support calls

Talk share, pace, longest monologue, interruptions and questions, plus a criterion-by-criterion scorecard with justification and the moment each one was met.

Built-in BANT and support standard scorecards, or your own criteria.
Roles per speaker with an AI suggestion you can override.
Objections with the response given, and next steps with owners.
Sentiment at the start, middle and end of the call.
Editable criteriaRegenerate after role changes
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
Roles changed — regenerate to update the score.
Subtitles & glossaries

Subtitles reviewed sentence by sentence

SRT and WebVTT with readability rules, up to four translated languages per recording, and glossaries that keep names and jargon spelled your way.

Up to 42 characters per line, 17 characters per second.
Status per sentence: OK, edited, needs review, outdated.
Only changed sentences are re-translated after a transcript edit.
Industry glossaries: IT, finance, legal, medical, construction, sales.
SRTWebVTTImport / export CSV glossaries
Subtitles · Chinese (Simplified)14/16 · 1 needs review
00:12Anna我们先从迁移状态开始。OK
00:19James第一阶段将在十月完成。Edited
00:27James我们需要检查 KSeF 中的发票。Needs review
00:41Anna我今天就发送摘要。Outdated
18 cues · 4 too fast — fix before export.

Keyboard-first in the editor

Ctrl + Space
Play / pause
Ctrl + F
Search & replace
Ctrl + T
Subtitles
Ctrl + I
AI analysis
Ctrl + Z
Undo, with a toast

Try it on a recording you already have

The free plan is enough to judge accuracy, speaker separation and the editor.

Start transcribing for free