Calculating and plotting performance metrics for speaker and biometric recognition made easy
Oxford Wave Research is proud to present the Bio-Metrics software for the Windows platform that makes evaluating the performance of your speaker or biometric recognition systems or algorithms easy and straightforward.
Have you spent hours trying to calculate and plot DET plots, Tippett plots, equal error graphs, etc. for your reports and papers?
Would you like to use graphical tools that are available for Windows to make your calculations easy?
Would you like to easily export your results to Microsoft Word, Powerpoint, etc?
Our Bio-Metrics software could be the right tool for you.
Our Bio-Metrics software provides the user with the capability to quickly plot the state of the art performance metrics such as DET plots, Equal Error Rate calculations, Tippett plots, etc, for speaker recognition or other biometric recognition software.
In order to do so it all it requires are the set of similarity scores obtained comparing features of the same individual as well the set of scores obtained comparing features of different individuals.
Read more about this here
January 2013 - SpectrumView Plus is now available on the appstore
Responding to all the feedback you have sent to us, we have now released a new powerful version of SpectrumView called SpectrumView Plus that incorporates a lot of your most-asked features.
*New* Frequency Pinch-to-Zoom – SpectrumView Plus supports a high-quality frequency zoom capability that allows you to zoom in frequency space to look at a specific tone or frequency in more detail. You can use the high resolution frequency zoom either vertically in the Spectrogram view or horizontally in the Spectrum Analyzer view.
*New* Recording and Playback of Audio Files - SpectrumView Plus is also a fully functional audio recorder that allows you to record, pause and play back files in high-quality uncompressed wave format with user-set sampling rates from 8000Hz up to 48,000Hz.
· *New* Tap Data Cursor - The tap functionality will allow you to tap on any part of the Spectrogram or Spectrum Analyzer display and obtain the exact frequency in Hertz (Hz) and level in decibels (dB) at that point. It is possible to do this either in a live audio stream or in a paused spectrogram.
*New* Musical Notes or Frequency Display on Tap : SpectrumView Plus will allow you to tap on anywhere on the Spectrogram or Spectrum Analyzer and obtain the cent-accurate musical note e.g. A4 (22), at that point. This feature is useful to tune your instruments or make sure you are hitting the right notes.
*New* Live Spectrogram and Spectrum Analyzer Views of Recorded Files : Files recorded using SpectrumView Plus can be played back and the Spectrogram and Spectrum Analyzer modes can be used to review the frequency space.
*New* Sharing Recorded Files by Email or iTunes File-sharing : The recorded audio files either be sent by email to a specified address (if they are under 10 MB) or transferred to your computer using iTunes. If you would like to transfer large files, SpectrumView Plus fully supports iTunes file-sharing which allows you to access your files through iTunes on your computer.
*New* High Quality Screenshots – This has been a much requested feature. It allows for taking high-resolution screenshots of only the Spectrogram or Spectrum Analyzer without the controls around it. The screenshot can be taken either in silent mode or with the camera click sound effect. The screenshots are directly stored on the device’s Camera Roll.
You can see a video walk-through of the new features here:
Visualize in real-time the frequencies that you can hear around you and even those you cannot. The SpectrumView analyser provides a high quality real-time spectrogram and spectrum analyser display with configurable sample rate and frequency resolution for the iPhone and iPad. You can use this application to measure the highest tones you can sing, to find the source of an annoying sound, for calibrating musical instruments or generally for all sorts of acoustic analysis.
Check out the screenshots (iPhone/iPad)
iPhone Screenshots - Real-time Spectrogram (Waterfall Display) and Spectrum Analyser
iPad Screenshots - Real-time Spectrogram (Waterfall Display) and Spectrum Analyser
The easy-to-use configurable interface provides the following features:
High quality real-time spectrogram view of audio measured at the inbuilt or external microphone to observe trends in the frequencies present in the recording environment
Real-time spectrum analyser that provides a sequence of snapshots of the frequencies present in recording environment
Multiple selectable sample rates of 8,000 Hz, 11025 Hz, 16,000 Hz, 22,050 Hz, and 24,000 Hz, 32,000 Hz, 44,100 Hz and 48,000 Hz.
Sampling rate of up to 48,000 Hz for up to 24,000 Hz measurements
High Frequency resolution with configurable signal analysis window lengths and Fast Fourier Transform (FFT) orders
Capability to pause or continue recording
Native support for portrait and landscape modes in iPhones and iPads
Sound intensity represented using a colour spectrum scale from -96 dBFS to 0 dBFS
Here's a slideshare of our presentation at the International Association of Forensic Phonetics and Acoustics (IAFPA) 2011. Slideshare does not allow audio to be embedded, so if you would like a copy of the audio, please send us an email.
Our paper titled, No, thank you, for the music': An application of audio fingerprinting and automatic music signal cancellation for
forensic audio enhancement has been accepted to The International Association of Forensic Phonetics and Acoustics 2011 Annual Conference 24-28 July, 2011; Vienna, Austria.
Oxford Wave Research Ltd is pleased to announce a strategic partnership with Patternlab, a Swiss research and development company, specialists in the applications of signal processing and pattern recognition/data mining techniques to real-world challenges.
Dr Krysztof Kryszczuk and Dr Jonas Richiardi from Patternlab, and Dr Anil Alexander from OWR, have worked together for close to ten years on a variety of algorithmic and pattern recognition problems. OWR looks forward to collaborating with Patternlab in finding solutions to new and exciting problems in the field of pattern recognition and biometrics.
