

Speech-to-text API
Supercharge your app with Banafo's topnotch automatic speech recognition engineWe trained our language models for over 60,000 hours.
Save yourself the hassle and easily integrate transcripts in your app.
Empower your software with voice to text functionality
The Banafo automatic speech recognition engine bridges the gap between spoken and written communication. If your application is already tapping into the increasing market of voice recording, now is the right time to grow your user base faster and easier than ever before.
Connect your software with our API in 3 steps and give your users access to automated transcripts of their calls, meetings, voice messages, web conferences, podcasts, videos…

How does the Banafo
speech-to-text API work?
Voice to text is a complicate matter and comes in many flavours. This flow helps you quickly understand how your application and our API can work together to unlock high quality transcripts for your users.

Your software works with voice recordings. Or your users can upload audio files within your application.
To protect the privacy of your users, the audio files and transcripts are exchanged over a secure connection.
All recordings in your Banafo account are protected from access by third parties. You can either keep the audio files, or have them automatically deleted when the transcript is ready.
Unlike many other transcription services, we developed, trained and constantly improve our own AI speech-to-text engine. We don’t rely on third party transcription services.The transcripts are generated on our servers, so you don’t need to invest in an expensive ICT infrastructure.
Integrate high-quality transcripts
in your app
We deliberately focused on the quality of transcripts over a pile of fancy features you probably don’t need, or that would require months of hard work to adapt your app.

Punctuation
Easy to read transcripts

Capitalization
On sentence level.

Filler words filter
Leaving out e.g.: um, uh, er, ah…

Timestamps
On phrase level

Speaker change detection
Coming soon

Channel transcription
Two channels speed up the transcription process

Output files
You can download an example further down the page

Languages
English German (coming soon) Dutch (coming soon)

Automatic language detection
Coming soon
Your voice recordings

Multi request
You can upload an unlimited amount of audio files

Delete recordings
Either manually or automatically when the transcript is generated

Accepted formats
.mp3, .ogg, .wav, .flac, .webm for best audio quality

Maximum file size
500 MB
Download
a transcript example
Click on a button below to download an example of a transcript file Banafo will send to your application.
You can configure the desired format in your API request.
Speech-to-text API pricing per minute
Although speech-to-text is a complex matter, we like to keep things as simple as possible for you when it comes to integrating transcripts… and pricing
No monthly subscription fee
US$0.01 /minute audio
No monthly minimum volume required
Up to 50% cheaper than most popular high-quality speech-to-text APIs
No setup fee
No surprises. You can see a detailed real-time overview in your account.
Estimate your monthly audio volume
Estimated transcript cost
Requirements to connect your app to the Banafo API
The API works independently from programming languages. The only requirement is that your application can deal with http requests.
Progressive Web Applications (PWA)
Android apps
iOS apps
Web applications

3 simple steps to integrate
speech-to-text in your software
It only takes 3 steps to connect your app with the Banafo transcription API
VIEW OUR API DOCUMENTATIONBROWSE OUR API FAQProvide your API Key x-api-key header in every http request
Contact us
Send all your questions to api@banafo.com or use the form below.
All fields are mandatory to direct your questions faster to the right team.