3 ways to get the most out of the Watson Speech to Text API
By Susan C. Daffron | 2 minute read | May 2, 2017
– Watson Speech to Text API converts audio voice into written text so you can add speech transcription capabilities to your applications.
– You can use it to create voice-controlled applications and customize the model to improve accuracy for the languages and content you care about.
– IBM offers a breadth of resources so you can quickly find what’s relevant to your app and business.
Watson Speech to Text converts audio voice into written text. Use Speech to Text to transcribe calls in a contact center for example to identify what is being discussed, when to escalate calls, and to understand content from multiple speakers.
To put it simply, you can use the Watson Speech to Text API to add speech transcription capabilities to your applications. The service has three interfaces you can use: a WebSocket interface, an HTTP REST interface, and an asynchronous HTTP interface. They have options to stream the audio or to send it as a single request. The service has a few extra features such as profanity filtering, formatting and word confidence as well.
You can also use Speech to Text to create voice-controlled applications – even customize the model to improve accuracy for the language and content you care about most such as product names, sensitive subjects, or names of individuals.
Like any feature-rich API, you don’t know what you don’t know about how to use it until you do a little digging. When you’re getting started with a new app, the trick is sifting through the information to determine what’s relevant to your app and what’s not. Fortunately, IBM has a lot of resources that can help.
1. See what Watson’s Speech to Text service can do
The Speech to Text service transcribes audio voice files into written text, so you can use it for a number of different types of applications. For example, you can use Speech to Text to transcribe calls or to create voice-controlled applications. You can also customize it to improve accuracy for a specific language or to understand specific content such as names of products or people. These links give you an overview of what the service can do:
- Speech to Text demonstration
- Watson sample apps are available on the Starter Kits page, including links to the Speech to Speech app and its associated code on GitHub.
- IBM Watson blog post “Getting robots to listen: Using Watson’s Speech to Text service” that shows how to use the service’s WebSocket interface with Python to extract speech from audio.
2. Learn how Speech to Text works
The Speech to Text service uses machine intelligence to combine information about grammar and language structure with knowledge about the composition of the audio signal to transcribe the human voice accurately. It continuously returns and retroactively updates the transcription as more speech is heard. To get started, you can:
- Read the Speech to Text documentation
- Run examples in the simple cURL-based tutorial
- Consult the API reference
3. Try it out for yourself
Once you start digging into code, you don’t have to go it alone. If you run into questions or have comments, you can share them in the Watson developerWorks forums. For more answers, you can check out these developer communities as well:
- The Watson forums on Stack Overflow.
- The Watson forum on dW Answers.
Get started with Speech to Text on Bluemix today.
See how our Watson Speech to Text API works with our free, 30-day Bluemix trial