Team D Speaker Recognition – the team that did the project sequence below
- www.data-compression.com/vq.html
- "vector quantization" "speaker verification" "text independent" – see pudn for lots of matlab files
- http://en.wikipedia.org/wiki/Speaker_recognition
- Matlab example of speaker recognition
- CPact Software – VoiceBox download
- VoiceBox Summary and download
- Speech Filing System – Mark Huckvale
- Theory on VQ Method
System 1: Luigi Rosa Code
- VQ method Luigi Rosa Voice Recognition Matlab P-Code – compiled ONLY – speakerrecognition.rar – How to donate to get the source code Solution based on Amin Koohi solution with improved distance metric
- Text-Independent Speaker Recognition Based on Neural Networks – problem: do not have: \Program
Files\MATLAB\R2008b\toolbox\nnet\nnet\nntrain\private\nn_train_feedback.m – This is from Matlab 2006a Neural Net ToolBox – MatLab Neural Net Version Map
- Speaker Recognition based on Relative Spectral Methodology – faster than VQ method but error prone when using same test / train files of approximately 2 seconds. What is the correct length of speech training file?
System 3: RASTA -PLP
- Dan Ellis -PLP and RASTA (and MFCC, and inversion) in Matlab using melfcc.m and invmelfcc.m – lots of matlab source files with related signal processing
Microsoft SAPI Based Solutions
- Planet source code examples SAPI usage to do control
- General SAPI setup / installation to work with end apps like Word
- Microsoft Speech API – works with VB.net – SAPI Book
- Microsoft Speech Recognition Engine – How to specify the SAPI training file
Gaussian Mixture Model (GMM) Hidden Markov Model (HMM)
-
AudioSeg open source project is a toolkit dedicated to audio segmentation and classification of audio streams. The toolkit implements standard reference algorithms such as energy-based silence detection, BIC segmentation and clustering as well as GMM/HMM classification. – Speaker Diarisation
Other Solutions
Sound and Wave handling via Visual Basic
- How to Control the Volume of Sound Files from Visual Basic
- Write a .wav file – Includes some info on how to do header
- Waveplayer using the waveOutWrite method – works with larger files
- Wave Class – Easily read and write RIFF
- TrueWaveEditor –
- How to encode a .wav file to MP3 using Windows DLL
Ancillary discoveries
- bar code reading from image
- Calling matlab from Visual Basic
- CELP – Code Excited Linear Prediction – most common cell phone speech compression method.
- Speex – open source CELP system
0 Comments