What is Watson Speech to Text?

Watson Speech to Text is a cloud-native solution that uses deep-learning AI algorithms to apply knowledge about grammar, language structure, and audio/voice signal composition to create customizable speech recognition for optimal text transcription.


Powerful real-time speech recognition

Transcribe your audio in real-time or via uploaded batch files using any of our available out-of-the-box language models, audio frequency options and transcription output features.

Value-add transcription features

Format and organize your transcripts as you need by using features such as speaker labels, smart formatting, keyword spotting, numeric redaction, word timestamps, confidence and alternatives.

Data control

Maintain control and ownership of your data with the assurance that your data is safe and secure. IBM will not collect, store or use your speech data without your explicit agreement and opt-in.


HansaWorld, improves global customer service with Watson by implementing a virtual assistant to help employees and clients interact directly with HansaWorld’s ERP solutions. When a user is connected to the talkbot, Watson Speech to Text service converts the audio stream of their voice into text, which is then fed into Watson Assistant to analyze the content of the inquiry, and respond with an appropriate answer or action.

Find out more

Business person speaking into cell phone
Two business people working with a laptop

Lingmo International

Lingmo International used Watson Speech to Text service to boost translation accuracy by 85% and to train models 50% faster to deliver smart translation solutions to businesses and consumers across voice and text platforms.

Find out more


To augment its call-centers, MRS BPO used IBM Watson technology to create a voice virtual agent, Adam, that customers can chat with over the phone.  Adam is a trailblazing conversational IVR that needs no human supervision in order to complete transactions, answer questions and redirect customers, with the help of Watson services.

Find out more


Consumer working at laptop and talking on cell phone

Get started now with Watson Speech to Text

By using our out-of-the-box language models, we give developers the tools to train and customize the service to learn the language of your business.