Continuous modeΒΆ

The REST API allows the use of continuous mode in speech recognition, but the complete audio must be sent in a single HTTP request. To use continuous mode, the license must be valid, and the parameter decoder.continuousMode=true. Further information in the section Continuous Mode.

An example of a request using the Linux curl command is shown below:

curl --header "Content-Type: audio/wav"  \
   --header "decoder.continuousMode: true" \
   --data-binary @<arquivo wav> \
   https://<HOST>:8025/asr-server/rest/recognize

Only after the audio has been sent is recognition finalized the final result returned. The recognition result follows the API standard, but contains the transcript of all the audio speech segments, as shown in the following example (certain fields have been omitted for the sake of simplicity):

../_images/continuousmode-rest.png