…unique recordings coming from different audio environments or even different times of the day, additional details can be analyzed leading to better results. Warning: Any human error in evaluation set preparation (in speaker uniqueness, placing recordings into wrong folder, etc.) affects the evaluation results, so it’s very important to prepare the data carefully. See SID Evaluation for more details…
Search: evalua
28 results
…when an error occurs, but view all errors and continue creating the evaluation set Fixed: SID Evaluator – invalid GID score values Fixed: SID Evaluator – missing SQE information in report Fixed: SID Evaluator – don’t save disabled recordings to evaluation set Phonexia Browser 3.40.3, BSAPI 3.40.4 (2021-05-28) Fixed: Some minor bugs in licensing system Phonexia Browser 3.40.2, BSAPI 3.40.2…
…this Microsoft Excel sheet demonstrating the sigmoid function: Score-to-Confidence. SID evaluation Before implementing Speaker Identification, it’s important to evaluate its accuracy using real data from the production environment. To evaluate the SID system, you’ll need enough of labeled data, i.e. recordings with speaker labels. The principle of SID system evaluation is to compare (voiceprints of) all the individual recordings against…
…Added the ability to define custom language in the speaker metadata Fixed: When discarding a changed photo, the confirmation dialog “Do you want to save…” popped up infinitely Fixed: Missing file names when the SID Evaluator evaluates speakers from the workspace Fixed: Unwanted extra comparisons when the SID Evaluator evaluates speakers from the workspace (instead of comparing only A x…
In this post, we break down the complexities of the initial installation process. By the end of the guide, you will be able to start processing your recordings with Phonexia Speech Technologies. 1. Download Evaluation package Download the Phonexia Evaluation package from https://partner.phonexia.com/kb/sp/speech-platform/evaluation-package/ Create a new directory and unzip the package into it in your desired location, for ex. C:/Phonexia/…
…to start processing your recordings with Phonexia Speech Technologies. 1. Download Evaluation package Download the Phonexia Evaluation package from https://partner.phonexia.com/kb/sp/speech-platform/evaluation-package/ Simply unzip the package to your desired location. Ideally avoid C:/Program Files as you may face issues later on with previleges 2. Save license.dat file Copy the license.dat file to the /SPE/ directory. Make sure the license.dat file is not…
…effect on the duration of comparison. It improves the system’s discriminability which decreases the Equal Error Rate (EER, for more information, have a look at SID Evaluation). Due to its easy use, it is highly recommended to use Mean Normalization in every SID implementation. False Acceptance Calibration Requirements: Number of audio recordings: 1000 divided by the desired False Acceptance Rate…
This part requires higher (and non-anonymous) access level.
How to solve this situation:
- Log in here if you are not logged in.
- Register here. It takes just a few clicks and it’s free.
…license contains records for all required modules. See Licensing article for additional information in FAQ Phonexia Browser, FAQ Speech Platform, FAQ Voice Inspector Permalink Q: What are the requirements for SID evaluation dataset? For evaluating the real life scenario of Phonexia Speaker Identification technology, the system needs to be calibrated by SID dataset. SID dataset (minimum requirements): To measure SID…
…FAQ Voice Inspector Permalink Q: What are the requirements for SID evaluation dataset? For evaluating the real life scenario of Phonexia Speaker Identification technology, the system needs to be calibrated by SID dataset. SID dataset (minimum requirements): To measure SID performance precisely, it’s important to prepare evaluation recordings set very carefully. The requirements are: 50+ known speakers, 200+ recordings in…
…REST clients connect to a single SPE then the X-Session must be shared between the REST clients; the session token must be shared and managed outside of SPE. If you are experiencing troubles during evaluation, set this directive to false and try Basic authentication as described here: https://download.phonexia.com/docs/spe/#Authentication. More advanced configuration and large scale deployment should be discussed within Phonexia…
Phonexia requires your acceptance of The End User Agreement before downloading, please check it. Step #1 – Download the package This package allows new users to try and evaluate Phonexia Voice Inspector. Phonexia Voice Inspector 5.2.0 for Linux 64-bit 281.9 MB Download Phonexia Voice Inspector 5.2.0 for Windows 64-bit 294.62 MB Download The package contains the following components, technologies &…
…client application on top of SPE (for technologies evaluation) This page summarizes major changes, bug fixes, and known bugs. We also announce important future plans and the end of life here. Major Changes: New Features and Fixes Speech Engine: Speech to Text (STT) We have several very interesting new features relevant to STT and KWS technologies. Both technologies are part…
This part requires higher (and non-anonymous) access level.
How to solve this situation:
- Log in here if you are not logged in.
- Register here. It takes just a few clicks and it’s free.
Step #1 – Download Try and evaluate all Phonexia speech technologies either via REST API using Speech Engine, or using the demo/testing GUI application named Phonexia Browser. Hardware requirements recommended: Intel Core i7 or better, 32 GB free RAM, 10+ GB storage (SSD preferred) minimum: Intel Core i5, 16 GB free RAM, 10 GB storage (SSD preferred) To prevent various…