?
Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH 2026
Academic editor: C. Busso, B. Sisman, B. Ramabhadran, J. Kim, P. Wagner
Under the general editorship: J. Wu
Chapters
Aparin G., Popov Vadim, Sadekova T. et al., , in: Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH 2026.: International Speech Communication Association, 2026. P. 631–640.
Whisper, a widely adopted ASR model, is known to suffer from hallucinations — coherent transcriptions generated for non-speech audio entirely disconnected from the input. We investigate whether hallucinations can be detected and mitigated through Whisper’s internal representations. We extract audio encoder activations and evaluate two representation spaces: raw Whisper activations and Sparse AutoEncoder (SAE) latents. ...
Added: October 6, 2026
Kuznetsov N., Kaledin M., , in: Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH 2026.: International Speech Communication Association, 2026. P. 4506–4510.
Audio Super-Resolution is a set of techniques aimed at high-quality estimation of the given signal as if it would be sampled with higher sample rate. Among suggested methods there are diffusion and flow models (which are considered slower), generative adversarial networks (which are considered faster), however both approaches are currently presented by high-parametric networks, requiring ...
Added: October 6, 2026
Language:
English
Aparin G., Popov Vadim, Sadekova T. et al., , in: Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH 2026.: International Speech Communication Association, 2026. P. 631–640.
Whisper, a widely adopted ASR model, is known to suffer from hallucinations — coherent transcriptions generated for non-speech audio entirely disconnected from the input. We investigate whether hallucinations can be detected and mitigated through Whisper’s internal representations. We extract audio encoder activations and evaluate two representation spaces: raw Whisper activations and Sparse AutoEncoder (SAE) latents. ...
Added: October 6, 2026
Savchenko V. V., L. V. Savchenko, Measurement Techniques 2025 Vol. 68 P. 453–463
The problem of non-invasive analysis of the vocal function of the speech apparatus based on the speaker’s
speech signal is addressed. A new method of acoustic analysis of a pulse-type voice source based on a two-stage
measurement procedure has been developed. The first stage of measurements provides for filtering of the voice
excitation signal of the vocal tract ...
Added: January 13, 2026
Portnova E., Butenko E., Sablin A. et al., , in: Interdisciplinary Research in Technology and Management 2024.: CRC Press Taylor and Francis, 2024.
Voicing game cues is one of the important tasks in game development. To avoid the need to record the voice of live
actors, neural network speech synthesis technologies can be used. Despite the rapid development of generative technologies, synthesized voices often lack the variety of intonations inherent in human speech. The problem with "one-to-many" is that ...
Added: November 23, 2025
Dmitrieva E., Maksim Kaledin, IEEE Signal Processing Letters 2025 Vol. 32 P. 3595–3599
Speech Enhancement techniques have become core technologies in mobile devices and voice software. Still, modern deep learning solutions often require high amount of computational resources what makes their usage on low-resource devices challenging. We present HiFi-Stream, an optimized version of recently published HiFi++ model. Our experiments demonstrate that HiFi-Stream saves most of the qualities of ...
Added: September 24, 2025
Tatiana Sherstinova, Nikolay Mikhaylovskiy, Evgenia Kolpashchikova et al., , in: Proceedings of the 35th Conference of Open Innovations Association FRUCT, 24-26 April 2024, Tampere, FinlandIssue 1.: FRUCT Oy, 2024. P. 253–258.
Contemporary advancements in NLP and neural network techniques are paving the way to enhance and harness traditional linguistic resources and corpora, as well as expand the methods of applying neural networks for complex language material. Thus, a weak point for both theoretical and applied linguistic tasks is the processing of spontaneous everyday speech. Two experiments ...
Added: November 29, 2024
Kharlamov A. A., Pantiukhin D., Borisov V. et al., Newcastle upon Tyne: Cambridge Scholars Publishing, 2024.
The monograph presents papers on the subject domain “Integral robot. Speech behavior”. These cover issues of a theoretical nature, including representation and processing of speech information in the human mind in the process of both text analysis and text generation, and specifically the need to use jointly working linguistic and extralinguistic models of the world, ...
Added: December 1, 2023
Shuranov E., / Series Computer Science "arxiv.org". 2021.
Text encodings from automatic speech recognition (ASR) transcripts and audio representations have shown promise in speech emotion recognition (SER) ever since. Yet, it is challenging to explain the effect of each information stream on the SER systems. Further, more clarification is required for analysing the impact of ASR's word error rate (WER) on linguistic emotion ...
Added: February 14, 2023
Andrey Malinin, Gales M., , in: Proceedings of the 9th International Conference on Learning Representations (ICLR 2021). ICLR, 2021.: ICLR, 2021. P. 1–31.
Added: November 1, 2021
Ryabinin M., Malinin A., Gales M., , in: Advances in Neural Information Processing Systems 34 (NeurIPS 2021).: Curran Associates, Inc., 2021. P. 6023–6035.
Added: October 31, 2021
Sokolov A., Savchenko A., , in: 2021 IEEE 19th World Symposium on Applied Machine Intelligence and Informatics (SAMI).: IEEE, 2021. P. 413–418.
This paper is focused on the finetuning of acoustic models for speaker adaptation goals on a given gender. We pretrained the Transformer baseline model on Librispeech-960 and conducted experiments with finetuning on the gender-specific test subsets. The obtained word error rate (WER) relatively to the baseline is up to 5% and 3% lower on male ...
Added: September 26, 2021
Meyer J., Rauchenstein L., Eisenberg J., , in: Proceedings of The 12th Language Resources and Evaluation ConferenceVol. 12.: European Language Resources Association (ELRA), 2020. P. 6462–6468.
We describe the creation of the Artie Bias Corpus, an English dataset of expert-validated <audio, transcript> pairs with demographic tags for age, gender, accent. We also release open software which may be used with the Artie Bias Corpus to detect demographic bias in Automatic Speech Recognition systems, and can be extended to other speech technologies. ...
Added: April 20, 2021
Sokolov A., / Series Computer Science "arxiv.org". 2021.
Text encodings from automatic speech recognition (ASR) transcripts and audio representations have shown promise in speech emotion recognition (SER) ever since. Yet, it is challenging to explain the effect of each information stream on the SER systems. Further, more clarification is required for analysing the impact of ASR's word error rate (WER) on linguistic emotion ...
Added: November 17, 2020
Sokolov A., Savchenko A., , in: 17th World Symposium on Applied Machine Intelligence and Informatics (SAMI).: IEEE, 2019. Ch. 19 P. 113–116.
In this article, we focus on the isolated voice command recognition for autonomous man-machine and intelligent robotic systems. We propose to create a grammar model for a small testing command set with self-loops for each state to return blank symbols for noise and out-of-vocabulary words. In addition, we use single arc connected beginning and ending ...
Added: October 21, 2019
Savchenko L.V., Savchenko A.V., Journal of Communications Technology and Electronics 2019 Vol. 64 No. 3 P. 238–244
In this paper, we studied the phonetic approach for voice processing. A method for automatic recognition
of speech signals, in which each quasistationary segment is associated with a fuzzy set of phonemes,
was developed. We proposed the operation of the probabilistic triangular norm for fuzzy sets corresponding
to the input frame and the nearest reference phoneme. The developed ...
Added: June 7, 2019
Mescheryakova E.I., Nesterenko L.V., , in: Computational Linguistics and Intellectual Technologies. International Conference "Dialogue 2017" ProceedingsVol. 1. Issue 16 (23).: M.: -, 2017.
Added: January 4, 2019