Search Results for: speech to text

Results 21 - 30 of 103 Page 3 of 11
Results per-page: 10 | 20 | 50 | 100

STT Language Model Customization tutorial

Relevance: 14%      Posted on: 2019-04-24

Language Model Customization tool (LMC) provides a way to improve the Speech To Text performance by creating customized language model. Language model is an important part of Phonexia Speech To Text. In a simplified way it can be imagined as a large dictionary with multiple statistics. The Speech To Text technology uses this dictionary and statistical model to convert audio signals into the proper text equivalents. Due to general diversity of spoken speech, the default generic language model may not acknowledge the importance of certain words over other words in certain situations. Language model customization is a way to inform the…

Language Identification (LID)

Relevance: 13%      Posted on: 2020-07-09

Phonexia Language Identification (LID) will help you distinguish the spoken language or dialect. It will enable your system to automatically route valuable calls to your experts in the given language or to send them to other software for analysis. Phonexia uses state-of-the-art language identification (LID) technology based on iVectors that were introduced by NIST (National Institute of Standards and Technology, USA) during the 2010 evaluations. The technology is independent on text and channel. This highly accurate technology uses the power of voice biometrics to automatically recognize spoken language. Application areas Preselecting multilingual sources and routing audio streams/files to language dependent…

How do you calculate SNR in Speech Quality Estimation?

Relevance: 12%      Posted on: 2019-07-01

Signal-to-Noise Ratio (SNR) is an important metric of whether a recording is worth further processing by other speech technologies, so it is part of our Speech Quality Estimation. However, calculating SNR automatically is not a trivial task. We use the fact that the statistical distribution of the frequencies in the waveform of speech has Gamma distribution. In contrast, noise has Gaussian distribution. So we can estimate the SNR by looking at the frequency distribution in individual frames. This approach to SNR estimation is based on the article by Kim Chanwoo, and Richard M. Stern, called "Robust Signal-to-Noise Ratio Estimation Based…

Speech Quality Estimation

Relevance: 12%      Posted on: 2018-04-02

Speech Quality Estimation (SQE) is a language-, domain- and channel-independent technology that quantifies the quality of an audio recording. 2 most important statistics used in the calculation of the SQE score are SNR (signal-to-noise ratio) and the bitrate of the recording. SQE is usually part of the rapid filtration process in deployments. SQE also measures over 20 other properties of the recording, all of which can be found in the output file and further processed. See description in SPE documentation. Typical use cases are: verification of recording quality on the input, searching based on quality of the recording, noise of…

What are STT preferred phrases and how to use them

Relevance: 12%      Posted on: 2020-11-26

Speech Engine version 3.32 and later includes new STT feature called Preferred phrases. This article explains what is the feature good for, how does it work internally and gives some tips for practical implementation. What are preferred phrases In the speech transcription tasks, there may be situations where similar sounding words get confused, e.g. "WiFi" vs. "HiFi", "route" vs. "root", "cell" vs. "sell", etc. Normally, the language model part of the Speech To Text does its job here and in the context of longer phrase or entire sentence prefers the correct word:  ×    I'm going to cell my car. Hmmm, such…

Sizing of the computing units for speech technologies

Relevance: 11%      Posted on: 2018-02-02

Best practices for good sizing of Phonexia technologies depend on a few facts: Intense work with large data sets requires good performance and bandwidth between RAM and CPU. It all depends on the size of the files with technological models data, usually loaded into RAM and used intensively for computing operations Always think only about physical cores of CPU (HT, VT features can't help in performance) Also seek for CPUs with a large L3 cache. And the better CPUs are those with higher l3_cache_size/#_of_physical_CPU_cores ratio. We currently assume that CPUs from the current Intel Xeon Family in the 4th generation…

How to configure STT realtime stream word detection parameters

Relevance: 10%      Posted on: 2020-03-28

One of the improvements implemented since Speech Engine 3.24 is neural-network based VAD, used for word- and segment detection. This article describes the segmenter configuration parameters and how they are affecting the realtime stream STT results. The default segmenter parametrs are as shown below: [vad.online_segmenter:SOnlineVoiceActivitySegmenterI] backward_extensions_length_ms=150 forward_extensions_length_ms=750 speech_threshold=0.5 Backward- and forward extension are intervals in miliseconds, which extend the part of the signal going to the decoder. Decoder is a component, which determines what a particular part of the signal contains (speech, silence, etc.). Based on that, decoder also decides whether segment has finished or not. Unlike in file processing…

Voice Inspector – supporting technologies

Relevance: 10%      Posted on: 2019-06-28

Automatic Speaker Identification (SID) is the most important but not the only Phonexia technology that is implemented in Voice Inspector (VIN). Apart from SID, forensic experts, users of VIN, can benefit from automatic Signal-to-Noise Ratio calculation, Voice Activity detection, Phoneme search, and a Wave editor which incorporates the waveform, spectrum and power panel. Let's have a look on how to utilize individual technologies. Signal-to-Noise Ratio Recording quality can strongly influence the reliability of SID results and so the outcome of a forensic case. Therefore, VIN uses a module of Phonexia Speech Quality Estimation (SQE) to calculate the Signal-to-Noise Ratio (SNR)…

Measuring of a software processing speed – what is the FtRT (Faster than Real Time)

Relevance: 8%      Posted on: 2019-10-30

Faster Than Real Time (FTRT) is metrics developed for defining software performance reference point. Using this metric you can collect "benchmark" data of real processing speed for reviewed software, which should be found - and reproduced - on exactly defined HW. Then, comparing various benchmarks result, you can compare performance of the specified software and its parts on different HW configurations. And vice versa - using the same metric you can compare software from different vendors on the same HW configuration and for the same processing task. We are recognizing two measurable metrics: Recording based FTRT is calculated from real…