BUBLEDUB / Product guides
Turn audio and video into text.
Search an interview, revisit a lecture or share meeting notes. Get text with timecodes, speaker labels and an AI summary.
BUBLEDUB Transcript converts an uploaded audio or video file into text without generating a dubbed voice. The reader connects timecodes to the recording, shows speaker labels and provides AI notes. You can add written translations and export a document or subtitle file.
- Transcription
- 1 credit / 5 source seconds
- Written translation
- +1 credit / 10 seconds / language
- AI notes
- Summary · key points · chapters
- Trial
- 150 credits ≈ 12½ min of transcription
How it works
Upload and choose Transcript
Select an audio or video file, then choose Transcript. You can request the original text alone or add written translations. No voice or framing settings are required.
Find the part you need
Open the completed transcript from the queue or library. Search a term, read speaker turns and use a timecode to return to the source.
Export for the next step
Choose TXT, DOCX, PDF, SRT or VTT. Where available, choose whether document exports include timecodes, speakers and notes.
Interviews, lectures and recorded meetings
For an interview, speaker labels help distinguish the questions from the answers. For a lecture, chapters and keywords help you return to a topic. For a recorded meeting, use the summary as a draft to check against the source before sharing it. BUBLEDUB processes files you upload; it is not a live meeting bot that joins calls.
A useful example: a one-hour interview
An original transcript of 3,600 source seconds costs 720 credits. One written translation adds 360, for 1,080 credits in total. That calculation uses source duration and does not include voice generation. If you only need the original transcript, do not add a target translation.
Choose the right export
TXT is convenient for plain text and copying. DOCX is useful when you want to edit in Word or another document editor. PDF is a fixed-layout document for reading or sharing. SRT and VTT carry timed subtitle cues for editors and players; they do not burn captions into a video by themselves. Font support can affect which PDF scripts are available.
Translate the text later
You can add written translations to a completed transcript without creating an audio dub. The studio supports up to ten translation targets in one round and shows the credit estimate before you proceed. The original text stays available alongside translations.
An example you can inspect
The landing page includes a small authored interview example with text, notes and a Spanish translation. It illustrates the reader and output format; it is not a customer recording or an accuracy benchmark. Download the text sample below to see the structure.
Check names, numbers and important conclusions
AI transcription, speaker attribution and summaries can make mistakes, especially with overlapping voices, noise, technical terms or proper names. Use timecodes to check uncertain passages against the recording. A generated summary should not be treated as a verbatim quote or a certified record.
Download the illustrative transcript (.TXT) ↗
Questions and answers
Is transcription the same as translation?
No. Transcription writes down speech in its original language. Translation changes the language. BUBLEDUB offers both, and charges written translations separately from the original transcript.
Does it work with MP3 and MP4?
Yes. Audio inputs include MP3, M4A, WAV, AAC, OGG, Opus and FLAC. Video inputs include MP4, MOV, WEBM and M4V. The browser must be able to read the chosen file.
Do I have to buy a subscription?
A subscription is optional. Web top-ups are one-off purchases, and monthly plans are available too. A new account starts with 150 free credits, enough for about 12½ minutes of transcription without written translation.