Occurrence of Keyword in recording, identified by technology with a given threshold.
Archives
Miss
Occurrence of Keyword in recording, not identified by technology with a given threshold.
False alarm
Occurrence of the word identified by technology as found Keyword, in a place that the Keyword was not actually said in the recording.
Hypotheses
List of found words on KWS output identified by technology as the variant for detection.
Threshold
Number defining how much the score of the found word must be to appear among detections.
Keyword pronunciation
Pronunciation of the keyword(s) is generated automatically (G2P, grapheme to phoneme) or produced from the lexicon of known words (“lexicon”) or converted from audio (phoneme transcription). It can be edited manually for each word (Phonexia do not limit the number of pronunciations per keywords/phrases).
Keyword list
Set of keywords or phrases defined together with keyword pronunciations (Phonexia do not limit the number of keyword lists).
Keyword
Word or a phrase that is searched by a user (defined by a user as an input for KWS technology). Phonexia does not limit the number of keywords in the keyword list. The higher number of keywords (500+) cause speed decrease.
Browser
Phonexia Browser is a powerful GUI tool for the Phonexia Speech Platform. It is designed for processing speech archives over Phonexia SPE.
WER
Word Error Rate – metrics for STT/LVCSR accuracy measurement
VP
Voice Print – output from spoken speech extraction process of SID. Unique mathematical representation of the specific speaker or recording is created in form of the iVector (for SID generation 3) or xVector (Deep Embeddings for SID generation 4).
VoIP
Voice over Internet Protocol – delivery of voice communications and multimedia sessions over Internet Protocol
VIN
Voice Inspector – Phonexia GUI application for forensic analysis
VATOOL
Voice Analytics Tool – predecessor of Phonexia Browser
VAD
Phonexia Voice Activity Detection
TANALYSIS
Command line version of time analysis extraction (TAE)
TAE
Time Analysis Extraction
STT
Phonexia Speech To Text, sometime also as Speech Transcription Technology (LVCSR based ASR technology)
SQE
Phonexia Speech Quality Estimator
SNR
SNR – signal-to-noise ratio
SPE
Phonexia Speech Engine (RESTfull API)
SIR
Speech Intelligence Resolver – predecessor of Phonexia Browser
SID
Phonexia Speaker Identification, multiple generations available marked by version like SIDv2 or SIDv3
SDK
Software Development Kit
SAL
Phonexia Speech Analytics (formerly named SPAS)
REST
REpresentational State Transfer (sometime also RESTful) application programming interface
PoC
Proof of the Concept
PHR
Phoneme recognizer – currently part of Keyword Spotting (Phonexia Keyword Spotting – acoustics based ASR, several tec…) technology in Phonexia Speech Engine (REST Application Program Interface)
Ph
Phoneme – The smallest phonetic unit in a language that is capable of conveying a distinction in meaning, as the m of mat and the b of bat in English.
OEM
Original Equipment Manufacturer