API Description
The Real-Time ASR API complies with the WebSocket protocol, and provides three modes: Streaming, Continuous, and Single-sentence.
The handshake request wss-URI of each mode differs, but the formats of requests and responses are the same.
Developers can use a WebSocket software package or library interface in the Java, Python, or C++ language to handshake with the Real-Time ASR engine, send the speech data, and obtain the recognition result. After the recognition work is done, close the WebSocket connection.
- For details about how to obtain the WebSocket handshake request wss-URI, see WebSocket Handshake Requests.
- For details about the format of a Real-Time ASR request, see Real-Time ASR Requests.
- For details about the format of a Real-Time ASR response, see Real-Time ASR Responses.
When a client uses the WebSocket protocol to access the RASR APIs, the connection can last 5 hours at most. If the WebSocket connection lasts more than 5 hours, the server automatically closes it.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot