Note! This page hasn't been finished The purpose of page is to share relted work or just problems and to track the meeting. Everyone is welcome to add paper or related ...
2014 May 27: ASR meeting (SJR, JD, FRM, MS, PS, JK, PJB) Systems we'll be working on over summer (coordinate them): BBC ( FRM to finish preprocessing LM data ...
This page is about the standardization of model training for ASR within CSTR. Next Meeting 4/11/2011 4pm (MeetingNov4th2011) Aims Generate Python libraries ...
RavichanderVipperla 05 Aug 2010 ATK Socket Interface The atk socket version is placed at: '/group/project/match/atkSocket' This setup works with the speech streaming ...
VolkerStrom 08 Jun 2007 Attaca (AuTomatic TArget Cost Algorithms) The full title of the project is "Automatic target cost and database design for unit selection speech ...
VolkerStrom 12 Jun 2007 Festival Development From the proposal: Festival will be enhanced (by Strom and Clark) to handle half phones as well as diphone units. We ...
Ongoing work This is the page where I will report on short term progress and ongiong work, while the pages under "Work Plan and Progress Reports" are intended to ...
VolkerStrom 13 Jun 2007 Automatic Target Cost and Database Design for Unit selection Speech Synthesis Project Summary The databases used in speech synthesis cannot ...
VolkerStrom 12 Jun 2007 Statistical analysis From the proposal: Replicate experiments of Beutnagel and Conkie on several existing Festival 2 voices. In the making ...
Main.simonk 15 Feb 2007 Barbara Forbes : articulatory feature recognition Goals Test a new feature representation by training ANNs to detect in speech, then compare ...
"Non linear time compression of clear and normal speech at high rates" Samples to support Interspeech 2015 submission Section 3 . Linear time compression Rate ...
MikeLincoln 28 Nov 2006 DemosAndVideos This page has been set up following the brainstorming session at the CSTR Meetings of 28/11/06 where CSTR demos were discussed ...
Main.dwang2 30 Sep 2007 30/09/07 In the past two weeks I mainly focus on two things. One is rearranging some results for the Speech Communication paper and ICASSP ...
Main.matthewa 07 Mar 2007 Dynamic Time Warping for Segmentation Just a quick over view of how to do this at the moment. Filenaming Files should have a naming comvention ...
RavichanderVipperla 10 May 2010 Components 1) Spherical microphone array with 32 microphones (EM32). Each microphone has programmable preamplifiers and A/D Convertors ...
Main.jyamagis 09 Nov 2007 Project Home 9th Nov 2007 Using CSTR nina's speech data, BIC values (strictly speaking, minus BIC) were calculated using Ergodic HMMs ...
Main.matthewa 27 Jul 2006 Project Home Letter to Sound Bibliography Overview One way of looking at emergent phones in this project is as a set of pronunciations ...
Main.matthewa 05 Jul 2006 Project Home Weekending 7th July 2006 Implemented phone/letter alignment algorith (as specified in Damper et al Aligning Letters and ...
Main.matthewa 06 Sep 2006 Project Home Pharaoh Scripts to build input for Phil Koehns MT system for doing LTS are in ephones/mt/scripts Results using top 20k frequent ...
Main.matthewa 18 Aug 2006 Project Home Evaluation of states v phones v silence Ergodic model with 45 states seeded with means from kmeans calculated over sppech regarded ...
F0 parametrisation using DCT coefficients First results 02.03.2010 The experiment goes like this: 1 Extract F0 using one of the methods used in HTS from the original ...
Use this page to track issues with the GlobalPhone corpus, i.e. things such as silent audio files, missing transcriptions etc. Some of this problems don't matter but ...
RavichanderVipperla 04 May 2010 The AMI online Recogniser Location : /group/project/ami10/onlineASR/online/releases/ASR.v09 To run the AMI recogniser, 1) Start ...
Main.s0566164 24 October 2008 Listen speech recognition meeting Please note: this wiki page is no longer being updated. Please see the new wiki page: https://www ...
Deep convex networks for ASR Imseng et al., Using out of language data to improve an under resourced speech recognizer (compares Tandem, KL HMM, SGMM), http: ...
2011 12Semester105.10.11 Language models (Steve) Model M Zweig and Chang, Personalising Model M, Interspeech 2011 Chen et al, Scaling shrinkage based LMs, Extended ...
2012 13Semester101.10.12Planning meeting 08.10.12 CANCELLED: Phone adaptive training for diarization(Mark) Mon P2c 05.pdf 15.10.12Feature space transforms with DNNs ...
2013 14Semester123.09.13Planning meeting30.09.13Semi Supervised Acoustic Model Training with Multi system Combination and Confidence Re calibration. (Huang et al) ...
2016 17Semester111.10.16Planning meeting18.10.16and semi supervised learning evaluation (Peter) 25.10.16 WaveNet: A Generative Model for Raw Audio (post with examples ...
Main.simonk 25 Sep 2006 2006 7Semester 1 2.10.06Progress reports 9.10.06 Songfang H. will lead the discussion of a generative topic model called Latent Dirichlet Allocation ...
Main.s0566164 21 Jan 2009 2008 9Semester 1 23.10.08Planning meeting 30.10.08Discussion of Interspeech papers on decision trees 1, 2 (Ravi and Peter) and log linear ...
2009 10Semester 107.10.09 Planning meeting13.10.09Matt Shannon's paper (1) on the Autoregressive HMM for more detail see (2)20.10.09Lin Bilmes Interspeech paper ...
Main.s0566164 13 Oct 2010 2010 11Semester 112.10.10 Review of Interspeech 2010 planning meeting19.10.10Recent work in speaker adaption using a VTLN prior Breslin ...
2011 12Semester218.01.12Planning meeting23.01.12 Discriminative training of long span LMs (Arnab) A. Rastrow, M. Dredze, S. Khudanpur, "Efficient Discriminative Training ...
Main.s0566164 01 Feb 2013 2012 13Semester204.02.13Minimum exact word error training (Rogier)11.02.13Planning meeting (review papers from SLT)18.02.13Convolutive non ...
2013 14Semester 213.01.14SVMs (Peter) For more details, see also the report20.01.14ASRU review planning27.01.14 KWS intro and review primary: sigir07.pdf further ...
Main.simonk 25 Sep 2006 2004 5Semester 2 18.1.05 Progress reports 25.1.05 Reading two Mark Gales SVM papers: N. Smith and M. Gales "Speech Recognition using SVMs ...
Main.simonk 25 Sep 2006 2005 6Semester 2 9.1.06Progress reports 16.1.06Auditory/CASA part 1: We will probably just cover the first of these in this meeting: Martin ...
Main.simonk 27 Oct 2006 All meetings are at 3pm in the CSTR meeting room, unless otherwise noted. 2006 7Semester 2 15.1.07Progress reports: Peter/Dong 22.1.07Songfang ...
Main.s0566164 23 Oct 2008 2007 8Semester 2 07.01.08No meeting 14.01.08discussion on the paper on beam forming by Seltzer and Stern 21.01.08discussions on ASRU2007 ...
Main.s0566164 18 Jan 2010 2009 10Semester 218.01.10 Review of ASRU 09 (Peter) Planning meeting25.01.10 Araki et al., NTT Japan, " DOA based speaker diarization system ...
2010 11Semester226.01.11Planning meeting02.02.11Lexical modelling: (1), (2). 09.02.11Tandem features for low resource languages: Thomas et al 16.02.11No meeting23 ...
Main.pbell1 07 Oct 2014 2014 15Term130.09.14Planning meeting06.10.14 Talk on LSTM language models (Daniel Renshaw) 13.10.14DNN regularisation Tomar and Rose, Li et ...
2014 15Term 2 12.01.15 SLT review and planning meeting19.01.15 large scale CTC trained RNN system review of sequence to sequence learning (Steve)26.01.15CD modelling ...
2014 15Term 3 18.05.14Planning meeting25.05.15 Noise Contrastive estimation (Siva) fast and simple algorithm for training neural probabilistic language models , exponential ...
Main.s0566164 23 Apr 2010 The following are items which we need to do (adapted from Junichi's email of 20/4/2010). Please edit the wiki to indicate when each task ...
Main.matthewa 04 Aug 2006 I think we are going to have fun with silence in this framework. If you have a look at the example alignment (after only 4 iterations over ...
Main.matthewa 27 Jul 2006 At present I've implements the Needleman Walsh algorithm for aligning letters to phones in the LTS system I've built. This is basically the ...
ML, PB reporting what they've done since last time Have been working on ROTK the central organisation tool for configuring modules etc got it to the stage of ...
QiongHu 07 Jun 2014 Neural network for speech recognition/synthesis Goals This page is for sharing tools , docs and your experiences for speech synthesis/recognition ...
Main.dwang2 30 Oct 2006 10.30 Need more paper reading for lattice matching and searching, possibly some techniques for confidence score estimation can be borrowed ...
Italicised figures are on the GlobalPhone dev set WER all other numbers are on the eval set WER unless otherwise stated. (hmm, strange things going on in ...
"Deep Neural Network based Postfilter for Statistical Parametric Speech Synthesis" Samples to support submission Section IV . Evaluation B. Context size of DNN ...
MikeLincoln 30 Jan 2007 Schedule Autumn 2006 26 Sep 2006 Introductory meeting followed by report from Steve R. and or Maria W. on the recent 'Preparing for FP7 ...
RavichanderVipperla 10 May 2010 10/5/2010 At this point, the architecture of the final system seems to be: 1 EigenMike connects to EMIB. 1 EMIB connects to ...
Psychoacoustic Models Lista workshop 2012: Using an intelligibility measure to create noise robust cepstral coefficients for HMM based speech synthesis Link for ...
Downsampling to 16k vs. training with 16k experiment The sentences in the next table are one of three types (not in this order): 1 HTS trained with 16k data ...
Spoofing and Anti Spoofing (SAS) corpus This is a temporary web page for the SAS corpus. The objective of the Spoofing and Anti Spoofing (SAS) corpus is a standard ...
Speak! speech synthesis meeting. The purpose of these regular informal meetings is to discuss and share progress relating to speech synthesis (audio and visual) research ...
Main.korin 02 Feb 2007 2007/2008 schedule Date Topic 25.10.07 Cancelled 01.11.07 Junichi Performance evaluation of HTS 2007 system 08.11.07 Visitors from Vienna ...
Main.korin 02 Feb 2007 Speak! speech synthesis meeting. The purpose of these regular informal meetings is to discuss and share progress relating to speech synthesis ...
Speak! speech synthesis meeting (2009 2010). The purpose of these regular informal meetings is to discuss and share progress relating to speech synthesis research ...
Speak! synthesis meetings schedule for 2012/2013 Semester 1 02.10.12 USTC visitor talks ( NOTE this will be in room IF 4.31/4.33 ): "Detection of synthesized ...
Details of suggested papers to read: Music to Dance Mappings: Ofli, Erzin, Yemez, Tekalp: Learn2Dance: Learning Statistical Music to Dance Mappings for Choreography ...
Details of suggested papers to read: Acoustic modelling etc: Chunwijitra, Nose Kobayashi. A speech parameter generation algorithm using local variance for HMM ...
Original signal (sampling frequency: 16kHz): Distortion: LPC synthesis with decreasing coefficient order order reduction (%)0102030405060708090 synthised speech ...
JunichiYamagishi 08 Nov 2006 Trajectory Modelling Meeting Trajectory modelling meetings are held on alternate Wednesay at 12:00 in the CSTR instrumented meeting ...
Welcome to the home of TWiki.CSTR . Ths page is for all CSTR related Wikis NB: There is a new wiki space!. You will need to use EASE credentials to access some pages ...
This is a subscription service to be automatically notified by e mail when topics change in this CSTR web. This is a convenient service, so you do not have to come ...
nop CSTR Web Preferences The following settings are web preferences of the CSTR web. These preferences overwrite the site level preferences in ., and can be ...