Speech Recognition

Article summary

Did you find this summary helpful?

Thank you for your feedback!

Feature	SR Http	SR Websocket
Accuracy	Better	A few percentages worse. Since this node works on partial speech fragments.
Performance	Better	Worse. Since this node needs to perform more work in order to produce several partial recognition results.
Responsiveness	Worse. Since this node waits until the current speech is completed until it starts actual recognition	Better, since this node Can produce partial results which can be visually presented to the customer.
Partial result support	No	Yes
Detailed result for each recognized word	Yes	No

Was this article helpful?

Table of contents