Speech to Text API

I am trying to create a windows application in C# windows application which should use IBM Watson service to perform a Speech to Text conversion.

How can we implement IBM Watson Speech to Text API service https://stream.watsonplatform.net/speech-to-text/api which should listen to User voice and write the output text.

I have searched many articles and in all those examples, .wav (audio) file was passed as input and the output text is written.

I am trying to perform this Speech to Text more real time instead of passing an existing or stored audio (.wav) file.

The same functionality can be done using Microsoft.Speech with code like recognizer.SetInputToDefaultAudioDevice(). Here the drawback is, we have to feed the DictationGrammer. So that the application could recognize the speech and convert to text.

Is it possible to do a similar task through Watson API Service as well?

---

The IBM Watson SDK for .NET has support for WebSockets which would satisfy your requirement of transcribing more real-time versus uploading an audio file:

WebSockets includes a single call that establishes a persistent connection with the service over the WebSocket protocol.

The SDK can be found here and the specific details for Speech to Text within the SDK can be found here.

链接地址: http://www.djcxy.com/p/64566.html

上一篇: 在plm中循环子集

下一篇: 语音到文本API