Your transcript is ready
No results. Try another word.
Export
Demo: nothing is uploaded or processed. Connect your API to make it real.
Transcribe audio to text
Speakers labeled, timestamps included, 40+ languages.
Upload an interview, a meeting or a podcast episode, or paste its RSS feed. Onda detects the language, tells the speakers apart and returns an editable transcript you can export in one click.
- Accepted formats
- MP3
- WAV
- M4A
- MP4
- RSS
- Export formats
- TXT
- DOCX
- SRT
- VTT
50 free credits a month · no card needed
Drop an audio or video file
MP3, WAV, M4A or MP4 · up to 2 hours
No file handy? Try a sample:
1 credit per minute of audio
How it works
-
Add your recording
Upload an MP3, WAV, M4A or MP4 file up to 2 hours long, or paste a podcast RSS feed and pick an episode.
-
Let Onda transcribe it
The language is detected automatically, each speaker is labeled and every paragraph gets a timestamp. A one-hour file is usually ready in a few minutes.
-
Edit and export
Fix names in the editor while the audio follows the text, then export to TXT, DOCX, SRT or VTT.
Key facts
- Input
- MP3, WAV, M4A, MP4 or an RSS feed
- Max length
- 2 hours per file
- Languages
- 40+, detected automatically
- Speakers
- Labeled automatically, names editable
- Export
- TXT, DOCX, SRT, VTT
- Turnaround
- A few minutes for a one-hour recording
Why use it
-
Built for conversations
Speaker changes, overlapping voices and filler words are handled, so an interview reads like an interview, not a wall of text.
-
Every word is searchable
Each word is time-coded: click it to hear the moment it was said, or search the transcript to find a quote in seconds.
-
Your own vocabulary
Add guest names, brands and jargon before transcribing so they are spelled right the first time.
Questions and answers
How accurate is the transcript?
With clear audio and one person speaking at a time, expect very few corrections. Background noise, crosstalk and strong accents lower accuracy, so run difficult recordings through Remove noise first.
Which languages are supported?
More than 40, including English, Spanish, French, German, Portuguese, Italian, Dutch, Arabic, Hindi, Japanese and Mandarin. The language is detected automatically, or you can set it yourself.
Can it tell speakers apart?
Yes. Each voice is labeled Speaker 1, Speaker 2 and so on. Rename a speaker once and the name updates throughout the transcript.
What if my recording is longer than 2 hours?
Split it into parts with Trim audio, then transcribe each part on its own.
What are the export formats for?
TXT and DOCX are for reading and editing; SRT and VTT are subtitle files with timestamps, ready for video players and for podcast apps that display transcripts.
Related guides
How to transcribe a podcast episode
Turn an episode into an accurate transcript with speaker names, fix the words software gets wrong, and export it as TXT, DOCX, SRT or VTT.
-
Subtitle video
Readable captions in minutes, burned in or as SRT.
Accepted formats: MP4 Export formats: SRT · VTT · MP4
-
Podcast show notes
Summary, chapters and quotes, drafted for you.
Accepted formats: MP3 · WAV · M4A · RSS Export formats: TXT · DOCX
-
Remove noise
Hum, hiss and echo gone. Your voice stays natural.
Accepted formats: MP3 · WAV · M4A · MP4 Export formats: WAV · MP3 · MP4
Try it now
Speakers labeled, timestamps included, 40+ languages.
50 free credits a month · no card needed