Confirming successful V‑Blaze deployment

The final steps in a V‑Blaze deployment is to restart the service and confirm that it is working. The following command will start the ASR engine service:

sudo service vociserver restart

There are multiple ways to confirm that a V‑Blaze instance is running:

  1. Check the err.log file using the following command:
    cat /var/log/vociserver/err.log 

    The contents of the file should be similar to the following:

    2020-07-24 17:53:46.809137 initializing 2020-07-24 17:58:10.457873 serving on port 17170
  2. Query the status of V‑Blaze with a V‑Blaze API request.
    curl http://vblaze_name:17171/sysinfo 
    Note: Use https:// if an SSL certificate has been installed with a load balancer like nginx in front of the ASR.

    The response should be similar to the following:

    { "sysinfo": { "aminfo": { "eng1-acoustic31": "3x GPU, 0.99 GiB", "eng1-acoustic31hs": "3x GPU, 0.56 GiB" }, "dbsize": "0", "expires": "2021-07-15 00:00:00", "gpu": [ "1x Tesla T4, 14.75 GiB" ], "gpudriver": "440.33.01", "gpumem": "4.67 GiB / 14.75 GiB", "licenses": { "ASRUse-voci_advtest-3c722804e93e4d0c": [ 21.1493, 0.0 ], "ASRWkr-voci_advtest-autosize-f6e81df": "permission" }, "maxconns": 350, "maxdecoders": 12, "maxstreams": 300, "product": "V-Blaze AWS : 300", "revdate": "2021-01-08 17:17:13", "uname": "uname", "uptime": "18:03:54 up 2 days, 18:30, 0 users, load average: 0.00, 0.01, 0.05", "version": "7.2.0-0", "webapi": "2.2.0-0" } }

    Refer to API Endpoints for detailed information on V‑Blaze API queries.

  3. Transcribe an audio file with a V‑Blaze API request:
    curl -F output=text -F file=@/opt/voci/server/examples/sample1.wav http://localhost:17171 /transcribe

    V‑Blaze should return a transcript of sample1.wav in plain text:

    The matter was resolved in a very professional manner. Your employees are very good.