Case Converter
Change text to UPPER, lower, Title, Sentence or other cases instantly.
Transcribe audio and video with speakers and timestamps, edit and search the transcript, and get AI summaries, meeting notes and action items.
The Soft Premium Tools AI Audio & Video Transcription tool converts meetings, interviews, lectures, podcasts, webinars and voice notes into an editable, searchable transcript with speaker labels and timestamps, then uses AI to write summaries, meeting notes, action items and chapters from that transcript.
Speech-to-text with punctuation and paragraphs, automatic language detection or 20+ languages to choose from, and an optional verbatim mode that keeps “um” and “uh”.
Different voices labelled Speaker 1, 2, 3 (rename them), and a time on every paragraph. Click a time to play the recording from there.
Correct any word, find and replace, undo, and restore the original. Search the transcript and summaries with highlighted matches.
Quick overviews, executive and detailed summaries, bullet points, chapters, lecture notes, podcast show notes, key takeaways, Q&A, or your own instructions.
Decisions, action items, owners and deadlines, but only when the recording states them, with suggested follow-ups kept separate. Open questions too.
Word (.docx), PDF, Markdown, text, SRT and WebVTT subtitles, CSV and JSON, with or without speaker names and timestamps.
Audio: MP3, WAV, M4A, AAC, FLAC, OGG and WebM. Video: MP4, MOV and WebM. Files are checked on our server, so renamed or damaged files are turned down with an explanation rather than charged.
A free account is needed. Each recording uses 1 export per 30 minutes (or part): a 45-minute meeting uses 2. You see the cost before you start, and it’s taken when transcription begins. Up to 20 AI summaries and all downloads for that recording are included. If transcription fails, the exports are returned. See pricing for export packs and monthly plans.
Recordings are processed by our speech-to-text provider, AssemblyAI; for summaries, only the transcript text is sent to Anthropic’s Claude. Your recordings, transcripts and summaries are visible only to you, and they’re deleted after 7 days or as soon as you delete them. Only upload recordings you have the right to use, and make sure the people in them know they may be transcribed: the rules on recording conversations depend on where you are. See our privacy policy.
Got a recording to write up? Choose the file above to see how long it is and what it costs before you start.
Transcribe a recordingEach recording uses 1 export per 30 minutes (or part), taken when transcription starts: from £1.00 for 5 exports, or included in a monthly plan. Its AI summaries (up to 20) and all downloads are included. If transcription fails, the exports are returned.
In browsers that support it, only the sound is uploaded, compressed on your device; video pictures stay on your device. The audio goes to our speech-to-text service, AssemblyAI. The transcript is kept in your account so you can come back to it, and the recording, transcript and summaries are deleted after 7 days, or straight away when you press Delete. For AI summaries, the transcript text (not the audio) is sent to Anthropic’s Claude. Please make sure you have permission to record and transcribe the people in your recording.
It labels different voices as Speaker 1, Speaker 2 and so on, which you can rename. It tells voices apart; it can’t identify who a person really is, and speakers who sound alike or talk over each other can be mixed up, so check important parts.
Clear recordings with one person speaking at a time usually come out very well; background noise, strong accents, crosstalk and specialist words make mistakes more likely. Passages the AI was less sure about are underlined so you can check them. Adding names and technical terms before you start helps.
The summaries are told to use only what’s in the transcript, to leave owners and deadlines as “not stated” unless they were said, and to mark suggested follow-ups separately from agreed actions. AI can still make mistakes, so check important points, names and dates: summary points link to the moment in the recording where possible.
MP3, WAV, M4A, AAC, FLAC, OGG, WebM, MP4 and MOV, up to 4 hours long and 500 MB. Longer recordings can be split into parts.
Yes. Download SRT or WebVTT subtitles, timed from the exact word timings. If you have edited the transcript, each subtitle uses the times of its edited paragraph instead.
Yes. It works in any modern browser on phones, tablets and computers, with nothing to install.
Change text to UPPER, lower, Title, Sentence or other cases instantly.
Count characters, letters, spaces and lines as you type.
Always freeCount words, sentences and paragraphs and see your reading time.
Always freeHear any text read aloud, then download it as an MP3 in a natural UK or international voice.