LIVE API REFERENCE

Vionis Flash
Detection API.

Classify Caller CH0 as synthetic or human using acoustic features and conversational timing from Agent CH1.

Connected to Vionis Flash

The website forwards validated audio to https://flash.vionis.site. Its /explain endpoint provides the verdict and model diagnostics. No simulated results are used by Voice Authenticator.

POST/detectaudio/wav or application/json
01 / REQUEST

Send a stereo call recording

Use 16-bit PCM WAV, 8–192 kHz, at least one second long. Caller is channel 0; agent is channel 1. Request bodies may be at most 25 MB, including base64 or multipart overhead. Flash resamples internally to 8 kHz; this website preserves the original bytes and channel order.

RAW WAV
curl https://vionis.site/detect \  -H "Content-Type: audio/wav" \  --data-binary @call.wav
JSON · POST /detect
{  "call_id": "call-001",  "audio_base64": "<base64 of the complete stereo PCM WAV>",  "sample_rate": 8000,  "channels": 2}

Replace audio_base64 with the complete WAV bytes, including its RIFF header. If supplied, sample_rate and channels must match the WAV. Multipart uploads use the audio or file field. Metadata does not change channel order. Server clients can call these routes without an API key; browser requests must come from this website.

02 / RESPONSE

Verdict and measured evidence

POST /detect returns the model’s boolean classification and confidence for that classification. The following is a schema example, not a result from your recording.

EXAMPLE · POST /detect
{  "is_synthetic": true,  "confidence": 0.87}

POST /explain and POST /api/detect return the same verdict plus measured diagnostics:

FieldMeaning
is_synthetic / confidenceCaller classification and its confidence, between 0 and 1.
p_synthetic / thresholdProbability of synthetic speech and the model decision threshold.
route / modelclean or fallback scoring route and the deployed model version.
line / featuresLine measurements and values used by the model. Missing measurements may be null or imputed.
replies_measuredNumber of caller replies measured against agent turns.
audioDuration in seconds, source sample rate, and two channels.
wall_ms / latency_msFlash processing time and server round-trip time, respectively.
subject_channel / channel_mapCaller 0 is assessed; agent 1 provides context.
mode / request_idlive and a unique request identifier.

A human prediction does not verify identity. Flash does not perform semantic verification. Very noisy audio, insufficient turns, or unfamiliar voices can reduce reliability.

03 / ERRORS

Errors never become a verdict

The website calls Flash /explain so decode failures cannot be mistaken for the raw /detect endpoint’s 50% fallback.

400
INVALID_JSON / INVALID_MULTIPART

Malformed request body.

403
CROSS_ORIGIN_REQUEST

Use this website or a server client.

413
AUDIO_TOO_LARGE

Reduce the complete request body to 25 MB or less.

415
UNSUPPORTED_MEDIA_TYPE

Use WAV, JSON, or multipart.

422
INVALID_AUDIO / INVALID_REQUEST / FLASH_AUDIO_REJECTED

Check the stereo PCM WAV header, channel mapping, duration, and metadata.

502 / 503
FLASH_UNAVAILABLE / INVALID_FLASH_RESPONSE

Flash is unreachable, busy, or returned an invalid result. Retry later.

504
FLASH_TIMEOUT

Flash did not finish within 60 seconds. Retry the recording.

GET /api/flash/health checks the live model service and returns HTTP 503 when unavailable.

04 / LIVE CONSOLE

Test your audio

Submit a real recording and inspect the exact response from the website’s Flash adapter.

Live request console
VIONIS FLASH
Your model response appears here.

Choose a WAV recording or supply its base64 bytes.

Send request uploads audio to Vionis Flash. The console does not save recordings.