URL: http://vblaze_name:17171/statusfull

Example Response:

       {
    "statusfull": {
        "#active": 0,
        "#conns": 1,
        "#done": 0,
        "#queued": 0,
        "#streams": 0,
        "efficiency": 100.0,
        "idlefor": 3327.259,
        "lasterror": "2021-11-11 11:34:07.880521",
        "lasterrormsg": "missing data WAVE header",
        "maxconns": 106,
        "maxdecoders": 56,
        "maxstreams": 56,
        "memfree": "99.33 GiB",
        "memhuge": [
            4801,
            13180
        ],
        "memlow": "79.64 GiB",
        "mlru": [
            [
                "eng1:callcenter",
                3327.6
            ],
            [
                "eng1:callcenter-xa",
                675514.9
            ]
        ],
        "responses": {
            "server": {
                "http-200": [
                    22,
                    "2021-11-18 11:42:43.428792"
                ]
            },
            "stream": {
                "broken pipe": [
                    3,
                    "2021-11-05 15:48:59.494302"
                ],
                "http-200": [
                    55,
                    "2021-11-11 12:10:47.499335"
                ],
                "http-400": [
                    11,
                    "2021-11-11 12:06:15.837620"
                ],
                "http-500": [
                    20,
                    "2021-11-11 11:34:07.881692"
                ]
            },
            "unknown": {
                "http-400": [
                    2,
                    "2021-11-10 15:41:26.845205"
                ],
                "http-408": [
                    1,
                    "2021-11-10 15:41:04.837708"
                ]
            }
        },
        "started": "2021-11-10 15:08:33.629982",
        "stream.audiosecs": 57782.5,
        "stream.count": 61,
        "stream.errors": 11,
        "stream.speed": 21.112,
        "stream.warnings": 5,
        "utilization.idle": 99.761,
        "utt.audiosecs": 28552.3,
        "utt.count": 6637,
        "utt.speed": 15.305,
        "utt.tat": 18.134,
        "utt.tat.max": 68.233
    }
}

Explanation:

The example response is a JSON object that shows detailed status information about the V‑Blaze engine.

The following list explains each key and value of a /statusfull query:

#active

Indicates the number of active transcription streams on V‑Blaze.

#conns

Indicates the number of ongoing connections to the V‑Blaze API. This includes connections to all endpoints, including /transcribe .

#done

Indicates the number of completed transcription streams that have not been deleted from the ASR engine.

#queued

Indicates the number of transcription streams waiting for the necessary resources to transcribe audio.

#streams

Indicates the number of current connections to the /transcribe endpoint.

efficiency

Indicates the efficiency of system between 0% and 100%. This is effectively current_speed / max_speed * 100

idlefor

Indicates the number of seconds since the last transcription.

lasterror

Indicates the time and date that the last error message was received.

lasterrormsg

Indicates the text of the last error message that was received.

maxconns

Indicates the maximum number of simultaneous connections to the V‑Blaze API. This includes connections to all endpoints, including /transcribe .

maxdecoders

Indicates the maximum number of audio decoders that can be used by V‑Blaze when processing audio files. Each decoder transcribes a single utterance at a time.

maxstreams

Indicates the maximum number of simultaneous connections to the /transcribe endpoint.

memfree

Indicates the amount of memory V‑Blaze will keep free.

memhuge

Provides insight into fragmentation of free memory by indicating the total size of free memory in the NORMAL zone.

Free memory is counted in huge fragments. In this context, huge is defined as either 512*pagesize or 1024*pagesize. The average pagesize is 4KiB, so the size of a huge fragment could range from 512*4KiB to 1024*4KiB.

memlow

Indicates the lowest reading of available memory observed since the last (re)start of the V‑Blaze server.

mlru

Standing for "models least recently used," mlru contains an entry for each loaded model. Each entry is an array where the first element is the model name and the second element is the number of seconds since that model was last used.

responses

A JSON dictionary mapping from endpoint types to details about all responses return by those endpoint types. There are three endpoint types:

• server - contains details for all responses from other endpoints (/status, /sysinfo, etc.)

• stream - contains details for all responses from the /transcribe endpoint

• unknown - contains details for all responses from unrecognized endpoints. For example, if someone attempted to access the endpoint /foo, which does not exist, the response for that request would be shown here

started

Provides a timestamp indicating the exact date and time V‑Blaze started.

stream.audiosecs

Indicates the total number of seconds of audio processed since V‑Blaze started.

stream.count

Indicates the total number of streams processed since V‑Blaze started.

stream.errors

Indicates the total number of erroneous streams since V‑Blaze started.

stream.speed

Indicates the current stream processing speed measured in audio seconds processed per second.

stream.warnings

Indicates the total number of streams with warnings since V‑Blaze started.

utilization.idle

Indicates the percentage of time the system has been idle since it was started. This numerical value is between 0.0 and 100.0.

utt.audiosecs

Indicates the total seconds of utterance audio processed since V‑Blaze started.

utt.count

Indicates the total number of utterances processed since V‑Blaze started.

utt.speed

Indicates the current utterance processing speed measured in audio seconds processed per second.

utt.tat

Indicates the average utterance turnaround time since V‑Blaze started, measured in seconds.

utt.tat.max

Indicates the maximum utterance turnaround time since V‑Blaze started, measured in seconds.