Important update

As we collectively overcome the impact of the COVID-19 pandemic, the safety and well-being of all attendees are the most important to us. Analyzing both the current situation and the projections over the next few months, the Organizing Committee has made the decision that “SpeD 2021” will take place as a fully virtual conference.
We will soon announce the methods in which presentations will be conducted.

SpeD 2021 – Welcome message

The “SpeD 2021” Organizing Committee invites you to attend the 11th Conference on Speech Technology and Human-Computer Dialogue, in Bucharest, Romania. Beyond its traditional topics, the “SpeD 2021” conference and international forum aims to highlight the current trends in AI-based speech technologies and applications, including deep learning algorithms and architectures, as well as the latest developments in machine learning for audio, image, and multimodal information processing, and human-robot interaction.
Furthermore, “SpeD 2021” encourages original contributions on how research in speech technology can provide real solutions for other pattern recognition and understanding applications.

The “SpeD 2021” Conference is organized under the Aegis of the Romanian Academy and is sponsored by IEEE and EURASIP. As was the case for all previous editions since 2009, the “SpeD 2021” Proceedings will be indexed by IEEE Xplore and by Thomson Conference Proceedings Citation Index.

The conference will proceed in the traditional face-to-face format in 2021 and will be held in Bucharest, at the Research Institute of the University POLITEHNICA of Bucharest “CAMPUS”. However, the Organizing Committee will be monitoring the COVID-19 situation over the next months, with complete concern for the health and safety of all participants. We will switch to a virtual conference if forced by the evolution of the COVID-19 pandemic. Authors will be notified on the definitive “SpeD 2021” format before the Paper Submission deadline.


  • Speech Analysis, Representation and Models
  • Spoken Language Recognition and Understanding
  • Computational Resource Constrained Speech Recognition
  • Text-to-Speech Synthesis
  • Speech-to-Speech Translation
  • Speaker Identification and Diarization
  • Speech Emotion Recognition
  • Speech Enhancement
  • Machine Learning and Deep Learning in Speech Technology
  • Natural Language Processing
  • Speech and Language Resources – Data Collection, Transcription and Annotation
  • Spoken Dialogue Systems
  • Speech Interface Technology, Design and Implementation
  • Spoken Document Retrieval and Text Mining
  • Speech Technology for Medicine and Psychology
  • Ethical Issues in Speech Technology Deployment
  • Forensic Speech and Audio Analysis
  • Speech Processing in Smart Homes and Assistive Technologies
  • Speech Disorders and Voice Pathologies (including the Context of the COVID-19 Pandemic) – Research and Services
  • Speech and Speaker Recognition Trends and Applications
  • Audio-Visual Speech Processing
  • Multimodal Information Retrieval
  • Audio Signal Processing and Applications
  • Intelligent Robots and Human-Robot Interaction
  • Machine Learning Applications for Embedded Systems
  • Deep Learning Algorithms, Architectures and Applications


  • Paper submission (5 – 6 pages, IEEE format): July 18, 2021 August 01, 2021 (final deadline).
  • Notification of acceptance and reviewers’ comments: September 12, 2021.
  • Submission of final papers: September 26, 2021.
  • Conference: October 13-15, 2021.