Turn a recording into text
A lecture, an interview, a conversation, a meeting nobody transcribed. Upload the file and get text with separated speakers, a summary and a list of decisions.
Test 10 minutes without an account or card. Free includes 120 minutes and 5 recordings a month; Pro is €9.99 or US$9.99 a month for 15 hours.
Three steps
You upload the file
Audio or video, up to 512 MB. No need to convert anything or trim the silence first.
We recognise the speech and separate the voices
The model transcribes the spoken language directly, and a separate step splits the recording by speaker — you can see where one person stops and another starts.
You get finished text
Transcript, summary and a list of decisions. Downloadable as PDF or Markdown, or sent to your inbox.
How this differs from a free browser upload
Voices are separated, not merged into one block
A two-person interview comes back as a dialogue, not as a wall of text where you have to guess who is speaking. Speakers are numbered; you assign the names yourself, one click for every line of that voice.
Speech is recognised directly, not translated first
English and Polish audio is transcribed in its original language, including inflection and proper names. It does not pass through another language on the way.
Not just text — a summary and decisions too
An hour of recording comes back with a paragraph of summary and a list of things to do. The raw transcript alone can be as long as the recording.
Search across everything you uploaded
Search the content of every recording at once, including word forms no dictionary knows. One recording is a file; ten are an archive that has to be searchable.
The recording stays in the European Union
We process the audio ourselves and pass it to nobody. Only the finished text leaves, and only so a summary can be written from it — the model provider is contractually barred from training on it.
Who uses this
- Research and thesis interviews — transcription you do not have to type out by hand.
- Recorded calls and meetings that had no bot on them.
- Lectures, training sessions and webinars — searchable text instead of two hours of video.
- A phone voice recorder: a note dictated in the car comes back as text with a task list.
Questions about recording transcription
How do I transcribe an interview?
Upload the audio or video recording. Talkenda turns the speech into text, separates the interviewees into distinct voices and prepares a summary. You can then assign names to the voices and download the result as PDF or Markdown.
What is the difference between transcription and a transcript?
Transcription is the process of turning a recording into text. A transcript is the finished text produced by that process, with utterances, timestamps and separated speakers.
Can both audio and video recordings be turned into text?
Yes. You can upload a typical file from a voice recorder, phone, online call, lecture or webinar. You do not have to extract the audio from a video or convert the file yourself first.
Does the free transcription separate speakers?
The 10-minute sample without an account gives you timestamped text, but no speaker separation or summary. A full upload on Free or a paid plan adds separated voices, a summary, tasks and recording history.
How much does transcription cost?
You can test up to 10 minutes for free, without an account. Free includes 120 minutes and up to 5 recordings a month with no card. Pro is €9.99 or US$9.99 a month for 15 hours of recordings.
How long does it take to transcribe a one-hour recording?
Usually up to 20 minutes from upload, including separated voices and the summary. You do not have to keep the page open — the finished text appears in your account.
And if the meeting has not happened yet
Then you do not have to record or upload anything — the bot joins the meeting in Meet, Teams or Zoom and does the same automatically. Where the platform provides captions, the transcript carries real names from the start.
See how the meeting bot worksData in the European Union · GDPR compliant · No training on your data