Frequently asked questions
Pricing, privacy, supported formats, languages, and how to get started.
Pricing
Do I need a subscription?
- No. convert.express is a pay-as-you-go service with no monthly plan and no commitment. You get 10 free minutes of transcription every day. Beyond that, you pay €0.02 per minute only for what you actually submit. There is nothing to cancel.
How much does it cost to transcribe a 1-hour file?
- A 60-minute file costs €1.20 (60 × €0.02). If you have already used your 10 free daily minutes, the full 60 minutes are billed; if not, the first 10 minutes are free and the remaining 50 cost €1.00. The minimum charge per payment transaction is €0.50.
How much does a 30-minute recording cost?
- A 30-minute recording costs €0.60. The first 10 minutes of each day are always free, so if you have not used your quota yet the cost is €0.40 for the remaining 20 minutes.
What happens after my 10 free daily minutes?
- After the free quota is used, each additional minute is billed at €0.02. You will be prompted to add a payment method if one is not already on file. The daily free quota resets at midnight UTC.
Is a credit card required to get started?
- No credit card is required to use the free 10 minutes. You only need to add a payment method when you want to transcribe audio beyond the daily free quota. Payments are handled by Stripe — card details are never stored on our servers.
Privacy & data
How long are my files stored?
- Audio files and transcripts are automatically deleted 24 hours after your last activity on the job. The deletion is unconditional — there is no way to recover a file after that window closes. Download your transcript before it expires.
Is my audio used to train AI models?
- No. Files sent to the transcription API you select — OpenAI Whisper, Microsoft MAI-Transcribe 1.5, or Google Gemini 3.5 Transcribe — are not used to train that provider's models under its API terms of service. convert.express does not use your audio, transcripts, or any personal data for advertising, profiling, or model training of any kind.
Who can see my uploaded recordings?
- Your audio is shared only with the transcription API you select — OpenAI Whisper, Microsoft MAI-Transcribe 1.5, or Google Gemini 3.5 Transcribe — and a temporary cloud storage provider. No human reviewer listens to your recordings, and no provider uses your audio for model training under their API terms.
Does convert.express use tracking cookies?
- No. The only cookies set are strictly necessary: an anonymous session token that links your browser to submitted jobs, and authentication cookies from Clerk if you sign in. No advertising, tracking, or analytics cookies are set without your consent.
Files & formats
What file formats are supported?
- Audio: MP3, WAV, M4A, FLAC. Video: MP4, MOV, MKV. Most common audio and video formats are accepted. The transcription engine extracts audio from video files automatically — there is no need to convert video to audio before uploading.
What is the maximum file size?
- Files up to 2 GB are supported. Long recordings are automatically split for processing and the transcript is merged into a single output. There is no separate limit on recording duration.
Can I paste a URL instead of uploading a file?
- Yes. The upload zone accepts YouTube links, Dropbox links, Google Drive links, and any direct MP3/MP4 URL. Paste the link into the URL field and convert.express will fetch and process the audio or video from that source.
Can I transcribe an MP4 video without converting it first?
- Yes. Upload the MP4 directly — the service extracts the audio track automatically. There is no need to extract audio to MP3 or WAV before transcribing. MOV and MKV video files are also accepted.
What output formats does the transcript come in?
- Transcripts are available as plain text (.txt), SRT subtitle files, and VTT subtitle files. You can download any format from the results page. If you only need the text without timecodes, use the plain text option.
Languages & models
Which languages are supported?
- 64 languages are supported across three AI models. OpenAI Whisper covers 57 languages including Arabic, Chinese, Dutch, French, German, Hindi, Japanese, Korean, Portuguese, Russian, Spanish, and Turkish. Microsoft MAI-Transcribe 1.5 covers 43 languages, and Google Gemini 3.5 Transcribe covers 60 — both add Assamese, Bengali, Gujarati, Malayalam, Odia, Punjabi, Telugu, which Whisper does not support. The language is detected automatically.
Do I need to specify the language before uploading?
- No. All three models detect the language automatically from the audio. You do not need to configure the language manually, though you can override it in the settings if you want to force a specific language.
Can I choose which AI model is used?
- Yes. You can select OpenAI Whisper, Microsoft MAI-Transcribe 1.5, or Google Gemini 3.5 Transcribe before submitting a job. Whisper supports the broadest general-availability status and includes single-pass translation to English at upload time. MAI-Transcribe 1.5 is fast and supports up to 200 entity-biasing phrases. Gemini 3.5 Transcribe has the widest language coverage (60 languages), in-model speaker diarization for up to 8 speakers, and up to 1,000 entity-biasing terms, but cannot translate. Every model's output can be translated afterward using the dashboard's Translate action, covered below.
Can it translate recordings into English or other languages?
- Yes, in two ways. Whisper can translate directly to English in a single pass at upload time, at no extra cost — Microsoft MAI-Transcribe 1.5 and Google Gemini 3.5 Transcribe do not offer this upload-time mode (translation is Not supported by Gemini at all). Separately, once any job finishes on any model, the dashboard's Translate action uses Claude AI to translate the transcript, summary, or action list into any of convert.express's supported languages, not just English.
What is entity biasing and which model supports it?
- Entity biasing lets you supply a list of words and phrases that the model should expect to encounter: interviewee names, product names, technical terms. This reduces errors on proper nouns. Microsoft MAI-Transcribe 1.5 supports up to 200 biasing phrases; Google Gemini 3.5 Transcribe supports up to 1,000 biasing terms (best results under 100). Whisper supports basic vocabulary hints.
Account & getting started
Do I need an account to try it?
- No. You can upload a file and get a transcript without creating an account or providing an email address. An anonymous session cookie links your browser tab to the job. Creating a free account adds a persistent daily 10-minute quota and a transcript history across sessions.
What does creating a free account add?
- A free account gives you a persistent daily 10-minute free quota that survives across browser sessions (without an account the quota resets with the session cookie), access to your transcript history, and the ability to manage your payment method for paid transcription.
Ready to transcribe?
10 free minutes every day. No account required.
Start transcribing →See also: pricing details · privacy policy · compare models · all 64 languages