Erinç Dikici

Erinç Dikici

Bogazici University
Electrical & Electronics Engineering
BUSIM Speech Processing Group

About Me

I am now working as a speech scientist at SAIL LABS Technology, Vienna, Austria.

I was a PhD candidate in the Department of Electrical and Electronics Engineering at Boğaziçi University, and a research assistant at Boğaziçi University Signal and Image Processing Laboratory (BUSIM), Speech Processing Group, working on statistical language modeling, speech recognition and speaker verification under the supervision of Murat Saraçlar. I received my M.Sc. degree in 2009 and my B.S. degree from the Department of Electronics and Communication Engineering of Istanbul Technical University, in 2006.

My current focus of research is discriminative language modeling for large vocabulary continuous speech recognition for Turkish. My master's research focuses on developing a robust speaker verification system for mobile and multimodal authentication applications. My other research interests include broadcast news segmentation, sliding text recognition (Video OCR), sign language recognition and music information retrieval.

Education

PhD

2009 - 2016

Bogazici University
Electrical and Electronics Engineering

MSc

2006 - 2009

Bogazici University
Electrical and Electronics Engineering

BSc

2002 - 2006

Istanbul Technical University
Telecommunication Engineering

Areas of Interest

  • Statistical language modeling
  • Speech recognition
  • Speaker recognition
  • Audio segmentation
  • Music information retrieval

Projects & Research

  • TMSC5x Digital Signal Processor programming
  • Digital Audio Broadcasting [Senior Project]
  • Sign Language Recognition
  • Musical Genre Classification
  • Music Transcription & Audio-Visual Note Conversion
  • Sliding Text Recognition
  • Musical Instrument Sound Classification & Synthesis
  • Speaker Clustering and Identification

Publications

Journal Paper

  1. E. Dikici, M. Semerci, M. Saraçlar, E. Alpaydın, “Classification and Ranking Approaches to Discriminative Language Modeling for ASR”, IEEE Transactions on Audio, Speech, and Language Processing, Vol. 21, No. 2, pp. 291-300, 2013. [web]

  2. M. Hruz, P. Campr, E. Dikici, A.A. Kındıroğlu, Z. Krnoul, A. Ronzhin, H.Sak, D. Schorno, H. Yalçın, L. Akarun, O. Aran, A. Karpov, M. Saraçlar, M. Zelezny, “Automatic Fingersign-to-Speech Translation System”, Journal on Multimodal User Interfaces, vol.4, no.2, pp.61-79, Springer, 2011. [web]

  3. O.Aran, İ.Arı, P.Campr, E.Dikici, M.Hruz, S.Parlak, L.Akarun, M.Saraçlar, “Speech and Sliding Text Aided Sign Retrieval from Hearing Impaired Sign News Videos”, Journal on Multimodal User Interfaces, 2008. [web]

In Proceedings - International

  1. E. Dikici, M. Saraçlar, “Unsupervised Training Methods for Discriminative Language Modeling”, 15th Annual Conference of the International Speech Communication Association (Interspeech 2014), 14-18 September 2014, Singapore. [pdf]

  2. E. Dikici, E. Prud’hommeaux, B. Roark, M. Saraçlar, “Investigation of MT-based ASR Confusion Models for Semi-Supervised Discriminative Language Modeling”, 14th Annual Conference of the International Speech Communication Association (Interspeech 2013), 25-29 August 2013, Lyon, France. [pdf]

  3. E. Dikici, A. Çelebi, M. Saraçlar, “Performance Comparison of Training Algorithms for Semi-Supervised Discriminative Language Modeling”, 13th Annual Conference of the International Speech Communication Association (Interspeech 2012), 9-13 September 2012, Portland, OR, USA. [pdf]

  4. A.Çelebi, H.Sak, E.Dikici, M.Saraçlar, M.Lehr, E.Prud’hommeaux, P.Xu, N.Glenn, D.Karakos, S.Khudanpur, B.Roark, K.Sagae, I.Shafran, D.Bikel, C.Callison-Burch, Y.Cao, K.Hall, E.Hasler, P.Koehn, A.Lopez, M.Post, D.Riley, “Semi-Supervised Discriminative Language Modeling for Turkish ASR”, 37th International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2012), 25-30 March 2012, Kyoto, Japan. [pdf]

  5. E. Dikici, M. Semerci, M. Saraçlar, E. Alpaydın, “Data Sampling and Dimensionality Reduction Approaches for Reranking ASR Outputs Using Discriminative Language Models”, 12th Annual Conference of the International Speech Communication Association (Interspeech 2011), 28-31 August 2011, Florence, Italy, pp. 1461-1464. [pdf]

  6. P.Campr, E.Dikici, M.Hruz, A.Kındıroğlu, Z.Krnoul, A.Ronzhin, H.Sak, D.Schorno, L.Akarun, O.Aran, A.Karpov, M.Saraclar, M.Zelezny, “Automatic Fingersign to Speech Translator”, eNTERFACE '10 Proceedings, Amsterdam, The Netherlands, 2010. [pdf]

  7. E. Dikici, M. Saraçlar, “Investigating the Effect of Training Data Partitioning for GMM Supervector Based Speaker Verification”, IEEE 24th International Symposium on Computer and Information Sciences (ISCIS 2009), Northern Cyprus, 2009. [pdf]

  8. O.Aran, İ.Arı, E.Dikici, S.Parlak, L.Akarun, M.Saraçlar, “Turkish Sign Language Dictionary”, IEEE Int. Conf. On Acoustics, Speech, and Signal Processing (ICASSP 2008) Show&Tell Exhibition, Las Vegas, NV, USA, 2008. [web]

  9. O. Aran, İ. Arı, P.Campr, E.Dikici, M.Hruz, D. Kahramaner, S. Parlak, L. Akarun, M. Saraçlar, "Speech and Sliding Text Aided Sign Retrieval from Hearing Impaired Sign News Videos", eNTERFACE '07 Proceedings, Istanbul, Turkey, 2007. [pdf]

In Proceedings - Turkish

  1. E.Dikici, M.Saraçlar, “Unsupervised Discriminative Language Model Training”, IEEE 22nd Signal Processing and Communications Applications Conference (SIU 2014), Trabzon, Turkey, 2014. [pdf in Turkish (with English Abstract)]

  2. E.Dikici, M.Saraçlar, “Curriculum Based Discriminative Language Model Training”, IEEE 21st Signal Processing and Communications Applications Conference (SIU 2013), Girne, North Cyprus, 2013. [pdf in Turkish (with English Abstract)]

  3. A.A.Kındıroğlu, B.E.Demiröz, H.Sak, E.Dikici, H.Yalçın, L.Akarun, M.Saraçlar, “Multimodal Kiosk for the Disabled”, IEEE 19th Signal Processing and Communications Applications Conference (SIU 2011), Antalya, Turkey, 2011. [pdf in Turkish (with English Abstract)]

  4. E. Dikici, M. Saraçlar, “Data Sampling Approaches for GMM Supervector Based Speaker Verification”, IEEE 19th Signal Processing and Communications Applications Conference (SIU 2011), Antalya, Turkey, 2011. [pdf in Turkish (with English Abstract)]

  5. E. Dikici, M. Saraçlar, “Data Duration and Model Size Dependency of GMM- and SVM-Based Speaker Verification Performance”, IEEE 17th Signal Processing and Communications Applications Conference (SIU 2009), Antalya, Turkey, 2009. [pdf in Turkish (with English Abstract)]

  6. E. Dikici, M. Saraçlar, “Sliding Text Recognition in Broadcast News”, IEEE 16th Signal Processing and Communications Applications Conference (SIU 2008), Didim, Turkey, 2008. [pdf in Turkish (with English Abstract)]

  7. H. Dibeklioğlu, E. Dikici, P. Santemiz, K. Balcı, L. Akarun, "Sign Language Motion Tracking and Generating 3D Motion Pieces Using 2D Features", IEEE 15th Signal Processing and Communications Applications Conference (SIU 2007), Eskişehir, Turkey, 2007. [pdf in Turkish (with English Abstract)]

  8. E. Dikici, "Musical Note Analysis Using Signal Processing Techniques", Müzik ve Bilim (6), 2006. [pdf]

MSc Thesis

"Effects of Data Duration, Model Size and Session Variability on Speaker Verification Performance", Bogazici University, Jan 2009. [pdf]
Supervisor: Asst. Prof. Murat Saraçlar

Contact Information

BUSIM - Signal and Image Processing Laboratory
Department of Electrical and Electronics Engineering
Bogazici University
34342 Bebek, Istanbul/TURKEY
Tel:+90 212 359 70 09
Fax:+90 212 287 24 65