The transcribed text is sent to Language Translator and the translated text is displayed and updated.The transcribed text from Speech to Text is displayed and updated.The audio is streamed to Speech to Text using a WebSocket.User presses the microphone button and captures the input audio.NOTE: This code pattern includes instructions for running Watson services on IBM Cloud or with the Watson API Kit on IBM Cloud Pak for Data.Ĭlick here for more information about IBM Cloud Pak for Data. Use React components and a Node.js server.Integrate Speech to Text, Language Translator, and Text to Speech in a web app.Retrieve and play audio from Speech to Text using a REST API.Use Language Translator with a REST API.Stream audio to Speech to Text using a WebSocket.When you have completed this code pattern, you will understand how to: When ready, selectĪ language and voice and then enable translation (and speech). Short sentences or utterances where a pause indicates a break.įor the best live experience, wear headphones to listen to the translated version of what your microphone is listening to.Īlternatively, you can use the toggle buttons to record and transcribe first without translating. To avoid backtracking or overlapping audio, only completed phrases are vocalized. You'll notice that the text is updated as words and phrases are completed and become better understood inĬontext. The best way to understand what is real-time transcription/translation vs. In your choice of locale-specific voices. Each completed phrase will be sent to Watson Text to Speech to be spoken Text will be displayed by the app in real time. Watson Language Translator service to be translated into the language you select.
As the input speech is transcribed, it will also be sent to a Built with React componentsĪnd a Node.js server, the app will capture audio input and stream it to a In this code pattern, we will create a language translator web app. Create a language translator app with voice input and output
0 kommentar(er)
