Skip to content

Is there any workaround way to process start up certification before I really start my recognition work #905

Closed
@alanywlee

Description

Dears,

I used Google speech API and found library will print two block of of message like certification
as following if I run java with option "-Dlog4j.debug=true".

Cipher suite mapping: TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256 => ECDHE-ECDSA-AES128-GCM-SHA256
Cipher suite mapping: SSL_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256 => ECDHE-ECDSA-AES128-GCM-SHA256
Cipher suite mapping: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256 => ECDHE-RSA-AES128-GCM-SHA256
...

I checked OUTBOUND time-stamp and noticed Google actually sent my speech usually 4.5 - 5.5 seconds after I called Google Speech API at the first time after application started in case of using 40~50 threads to run simultaneously.

2017-10-31 16:54:14,432 [DEBUG] [er-ELG-2-1] 
----------------OUTBOUND--------------------
[id: 0xe75ba63b, L:/10.137.35.174:57240 - R:speech.googleapis.com/216.58.200.42:443] SETTINGS: ack=false, settings={ENABLE_PUSH=0, MAX_CONCURRENT_STREAMS=0, INITIAL_WINDOW_SIZE=1048576, MAX_HEADER_LIST_SIZE=8192}
------------------------------------

My problem is the first batch of jobs always work much longer than the following batch is pretty annoying to measure recognition efficiency. Is there any workaround way to process the start up certification before I really start my recognition work.

Any suggestion is highly appreciated.
Best regards,

Alan

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions