Search Results for: language model customization

Results 21 - 30 of 57 Page 3 of 6
Results per-page: 10 | 20 | 50 | 100

Terminology

Relevance: 6%      Posted on: 2017-06-15

Document which briefly describes processes and relations in Phonexia Technologies with consideration on correct word usage.   SID - Speaker Identification Technology (about SID technology) which recognize the speaker in the audio based on the input data (usually database of voiceprints). XL3, L3,L2,S2 - Technology models of SID. Speaker enrollment - Process, where the speaker model is created (usually new record in the voiceprint database). Speaker model: 1/ should reach recommended minimums (net speech, audio quality), 2/ should be made with more net speech and thus be more robust. The test recordings (payload) are then compared to the model (see…

Age Estimation

Relevance: 6%      Posted on: 2018-04-12

Phonexia Age Estimation (AGE) estimates the age of a speaker from audio recording. The process of voiceprint extraction is similar to the extraction of SID, but as a result different features get extracted; therefore, the voiceprints extracted from AGE and SID are not mutually compatible. Technology Trained with emphasis on spontaneous telephony conversation The technology is language-, accent-, text-, and channel- independent Compatibility with the widest range of audio sources possible (applies channel compensation techniques): GSM/CDMA, 3G, VoIP, landlines, etc. Input Input format for processing: WAV or RAW (8 or 16 bits linear coding), A-law or Mu-law, PCM, 8kHz+ sampling…

LM

Relevance: 6%      Posted on: 2018-02-01

Language Model (“vocabulary” in STT technology)

Keyword Spotting results explained

Relevance: 6%      Posted on: 2019-06-12

This article aims on giving more details about Keyword Spotting outputs and hints on how to tailor Keyword Spotting to suit best your needs. Scoring and results explanation Keyword Spotting works by calculating likelihoods that at a given spot occurs a keyword or just any other speech, and comparing those two likelihoods. The following scheme shows Background model for anything before the keyword (1), the Keyword model (2) and a Background model of any speech parallel with the keyword model (3). Models 2 and 3 produce two likelihoods – Lkw and Lbg (any speech = background). Raw score is calculated…

SPE3 – Releases and Changelogs

Relevance: 6%      Posted on: 2019-10-02

Speech Engine (SPE) is developed as RESTfull API on top of Phonexia BSAPI. SPE was formerly known as BSAPI-rest (up to v2.x) or as Phonexia Server (up to v3.2.x). This page lists changes in SPE releases. Releases Changelogs == SPE v3.18.x == Speech Engine 3.18.2 (10/14/2019) - DB v1300, BSAPI 3.22.1 Fixed: Customized STT model fails on Windows with Request for next state but ending state reached. error message Speech Engine 3.18.1 (10/01/2019) - DB v1300, BSAPI 3.22.0 New: DICTATE technology has been renamed to STT_STREAM (/technologies/dictate -> /technologies/stt/stream) (for backward compatibility, the /technologies/dictate endpoint is internally redirected) New: SID/SID4…

LP

Relevance: 3%      Posted on: 2018-02-01

Language Print - output data from LID technology

Voice Activity Detection

Relevance: 3%      Posted on: 2018-04-02

Voice Activity Detection is a language-, domain- and channel-independent technology that identifies parts of audio recordings with speech content vs. non-speech content. It creates labels for speech and other signals in the recording; this can then serve as a decision point whether to process the recording by other technologies or not. VAD is usually part of rapid filtration process in deployment. Typical use cases are: detection of present or absent human speech for voice processing, filtering non-speech parts of the recording, filtering out recordings with not enough net speech to be processed by other technologies voice activated process, etc. The…

Speaker Diarization (DIAR)

Relevance: 3%      Posted on: 2017-06-26

About DIAR Phonexia Speaker Diarization (DIAR) enables segmentation of voices in one monochannel audio record. Technology Trained with emphasis on spontaneous telephony conversation The technology is language-, accent-, text-, and channel- independent Compatibility with the widest range of audio sources possible (applies channel compensation techniques): GSM/CDMA, 3G, VoIP, landlines, etc. Input Input format for processing: WAV or RAW (8 or 16 bits linear coding), A-law or Mu-law, PCM, 8kHz+ sampling Output Log file with processed information (segmentation of speech, silence, and technical signals – ie. elimination of phone lines beeps, DTMF tones, music, pauses, etc.) Audio file extracted for each…

LPA

Relevance: 3%      Posted on: 2018-02-01

Language Print Archive - pack of language prints from the recordings spoken in the same language/dialect. Used for the language identification in LID comparison.

Voice Inspector – Interpretation of results

Relevance: 3%      Posted on: 2019-06-24

Introduction Phonexia Voice Inspector (VIN) is a tool for forensic automatic speaker identification, compliant with the Methodological Guidelines for Best Practice in Forensic Semiautomatic and Automatic Speaker Recognition, published by the European Network of Forensic Science Institutes.  This post explains individual SID score types and ways to visualize the results in a speaker identification case implemented in Voice Inspector. Evidence In VIN, the term evidence has two meanings. In general, it refers to any SID score that the system calculates for any pair of recordings in the case. These scores are the output of the Phonexia SID technology which runs…