AI Transcription: Convert Audio to Text Online
VdoBloom’s AI transcription tool converts an audio recording into written text in your browser. Upload a voice memo, interview, podcast episode, lecture or voiceover, and the speech is transcribed in the cloud with the ElevenLabs speech-to-text model. The transcript appears on the page with a one-click copy button, ready for notes, show notes, blog posts, scripts or subtitles.
Is it free? New accounts get 10 free credits, which covers one minute of audio (6 credits). After that, transcription costs 6 credits per started minute, measured from the file itself, and the exact cost is shown before you click Transcribe. The Lite pack gives 300 credits for $19.50 one-time (credits never expire) or $15 a month, enough for 50 minutes of audio.
There is nothing to install and no language to pick: the spoken language is detected automatically. If a file cannot be read or the provider fails, the credits are refunded.
How it works
- 1
Upload your audio file
Choose a recording from your device. MP3, WAV and M4A work best. The tool reads the file’s duration as soon as you select it.
- 2
Check the cost
The credit cost for that length (6 credits per started minute) is shown before you start, so you can compare it with your balance.
- 3
Transcribe
Click Transcribe Audio. The file is processed in the cloud; longer recordings take longer, and you can keep the tab open while it works.
- 4
Copy and reuse the text
The transcript appears on the page. Copy it to your clipboard and paste it into a document, caption editor, CMS or notes app.
Why use this tool
Automatic language detection
You do not need to choose a language before uploading. The model identifies the spoken language and transcribes in it.
Fair, measured billing
Cost is based on the real length of the file, rounded up to the next minute, and shown before you start. Failed jobs are refunded.
Works in the browser
No app, plugin or desktop software. Upload from a laptop or phone and copy the result from the same page.
Part of a full audio and video suite
Turn a transcript back into a clean AI voiceover, or add word-by-word captions to a video, without leaving VdoBloom.
How much does AI transcription cost?
Transcription is billed per started minute of audio at 6 credits per minute. The length is measured from the uploaded file on the server, so a 61-second clip counts as two minutes. The cost is displayed in the tool before you start, and you are only charged once the file has been accepted.
With the Lite pack (300 credits for $19.50 one-time, or $15 a month), you can transcribe 50 minutes of audio. One-time credits never expire; subscription credits reset each billing cycle.
| Recording | Length | Minutes billed | Credits |
|---|---|---|---|
| Voice memo | 0:45 | 1 | 6 |
| Short interview clip | 2:30 | 3 | 18 |
| Podcast segment | 10:00 | 10 | 60 |
| Lecture or webinar | 45:00 | 45 | 270 |
How AI transcription works
When you upload a file, VdoBloom first measures its real length on the server, which is what the price is based on. The audio is then sent to the ElevenLabs speech-to-text model, which listens to the whole recording, works out which language is being spoken and writes out the words with punctuation. When the job finishes, the text is returned to the page.
Because the model hears the full recording rather than single words in isolation, it uses context to pick between similar-sounding words, which is why clear, continuous speech gives the cleanest result. Short recordings usually finish quickly; an hour-long lecture takes longer, so leave the tab open until the transcript appears.
Who uses audio to text, and for what
Podcasters paste the transcript into their show notes and pull quotes for social posts. Students and researchers transcribe lectures and interviews so they can search, quote and summarise them. Journalists turn recorded calls into text they can check against their notes. Creators transcribe a rambling voice memo, tidy it into a script and then voice it with AI Text to Speech.
Marketers use transcripts of webinars and customer calls as raw material for blog posts, FAQs and ad copy, since spoken explanations often contain the exact words customers search for.
Turning a raw transcript into usable text
A transcript is a faithful record of what was said, including false starts and filler words. A short editing pass makes it publishable:
- Add speaker names before each change of speaker in interviews.
- Remove filler words (“um”, “you know”) unless you need a verbatim record.
- Break long passages into paragraphs at each new topic.
- Add headings if the text will become a blog post or show notes.
- Check every name, number, date and product term against your source.
What you can transcribe
The tool accepts audio files. If your recording is inside a video, export the audio track as MP3 or WAV first, or use AI Captions to subtitle the video directly.
- Interviews and research calls
- Podcast episodes, for show notes and blog posts
- Lectures, classes and webinars
- Voice memos and dictated drafts
- Voiceovers and scripts you want as editable text
- Meeting recordings exported as audio
Tips for a more accurate transcript
AI transcription accuracy depends mostly on the recording. These habits help:
- Record close to the microphone in a quiet room; background music and echo cause errors.
- Keep one person speaking at a time where possible.
- Trim long silences or unrelated sections before uploading to save credits.
- If a file is rejected, re-export it as MP3 or WAV and check it is not empty or corrupted.
- Proofread names, brand terms and numbers before publishing the text.
What this tool does not do (yet)
To set expectations before you spend credits:
- The result is plain text with a copy button. There is no SRT, VTT, DOCX or PDF export and no timestamps on this page.
- Speaker names are not labelled in the transcript.
- Upload only: there is no live recording, and no transcription from YouTube or other links.
- Video files are not accepted; upload the audio track instead.
- Files can be up to 200 MB.
Read more: Need timed subtitles? Use AI Captions · How to add captions to a video with AI
From transcript to finished content
A transcript is often the first step. Clean it up, then paste it into AI Text to Speech for a polished voiceover in one of 36 voices, or use it as the script for a short video. Podcasters can pull quotable moments from the text and turn them into vertical clips.
Read more: AI Text to Speech · Turn a podcast into short clips
Frequently asked questions
Is there a free AI transcription option?
Yes. Every new VdoBloom account gets 10 free credits, enough to transcribe one minute of audio (6 credits). After that, transcription costs 6 credits per started minute. The Lite pack’s 300 credits ($19.50 one-time) cover 50 minutes.
How do I convert audio to text with AI?
Open the Audio to Text tool, upload your recording (MP3, WAV or M4A work best), check the credit cost shown for its length, and click Transcribe Audio. The transcript appears on the page, and the copy button puts it on your clipboard.
How accurate is AI transcription?
Accuracy depends mainly on audio quality. Clear speech recorded close to the microphone transcribes well; heavy background noise, music, overlapping speakers and strong echo reduce accuracy. Always proofread names, numbers and technical terms before publishing.
Which languages can it transcribe?
You do not choose a language. The ElevenLabs speech-to-text model detects the spoken language automatically and transcribes in that language, so recordings in Hindi, Spanish, French and many other languages work without changing a setting.
How much does it cost to transcribe one hour of audio?
One hour is 60 minutes at 6 credits per minute, so 360 credits. Cost is rounded up per started minute and is always shown before you start.
What audio formats are supported?
The tool accepts audio files, and MP3, WAV and M4A are the most reliable. Files can be up to 200 MB. Video files are not accepted, so export the audio track first.
Can I get subtitles (SRT) from my audio?
Not from this page, which returns plain text. To add timed captions to a video, use VdoBloom’s AI Captions tool, which generates word-by-word subtitles on the video itself.
Does it identify different speakers?
The transcript is returned as plain text without speaker labels. For interviews, you can add speaker names while proofreading.
What happens if the transcription fails?
If the file cannot be read or the provider is unavailable, the credits for that job are refunded and you see a message explaining what to try, such as re-exporting the file as MP3 or WAV.
Can I turn the transcript back into speech?
Yes. Paste the cleaned-up text into VdoBloom’s AI Text to Speech tool to get a voiceover in one of 36 Gemini and xAI voices, at 6 credits per 1,000 characters.