NEWSLETTER
Apple sponsors the IEEE International Conference on acoustic, speech processing and signals (ICASSP), which takes place in person from April 6 to 11 in Hyderabad, India. ICASSP is a conference in the field of signal processing and its applications. Below is the workshop schedule and events sponsored by Apple in ICASSP 2025.
Schedule
Apple stand from April 6 to 11 from 09:00 to 17:00 in Booth C3 at the International Convention Center of Hyderabad. All time listed are in GMT +5: 30.
Monday, April 7
Wednesday April 9
- POSTER
- Recovery correction of the voice recognition errors of entity appointed
- 11:30 – 13:00, advances in LLMS and E2E architectures for Asr, room poster 2f
- Ernest Pusteri, Anirudh Kashi, Bortik Bandyopadhyay, Nadia Hyder, Sayanan Mahainder, Ravitaja Anantha, Daben Liu (Capital One), Sashank Gondala (more ai)
- Network lunch
- Women in signal processing (WISP)
- 13:00 – 15:00
- Location: HICC Architecture Room
- Ahmed Tewfik and his colleagues will represent Apple in women's lunch in signal processing.
- Industry Forum Talk
- ICASP 2025 Industry Forum
- 16:30 – 16:45
- Location: Hall Industry Forum
- The ai era on the device: Apple's bioacoustic systems for sound and voice recognition
- Venkataramanan Subramanian
- POSTER
- Speaker-ILP: Learning not supervised of speaker characteristics with pseudo-labels based on vector I
- 17:00 – 18:30, recognition of the speaker II, poster of the 2d room
- Shinji Watanabe (Carnegie Mellon University), Jeeweon Jung (Carnegie Mellon University), Ahmed Hussen Abdelaziz, Takuya Higuchi, Zak Aldeneh, Li-Wei Chen (Carnegie Mellon University), Stephen Shum, Tatian Likomomanenko, Barry-Jame Obry-Ju
Thursday April 10
- Oral presentation
- Compact Neural TTS voices for accessibility
- 09:15 – 09:25, TTS ARCHITECTURE, MRG. 04
- Kunal Jain, Eoin Murphy, Deepanshu Gupta, Jonathan Dyke, Saumya Hiren Shah, Vasileios Tsiaras, Petko Petkov, Alistair Conkie
Accepted documents
An efficient and transmission visual loudspeakers detection system
Arnav Kundu, Yanzi Jin, Max Horton, Mohammad Sekhawat, Danny Tormoen, Devang Naik
Compact Neural TTS voices for accessibility
Kunal Jain, Eoin Murphy, Deepanshu Gupta, Jonathan Dyke, Saumya Hiren Shah, Vasileios Tsiaras, Petko Petkov, Alistair Conkie
A ASR contextualization with LLM using phonetic recovery increase
Zhihong Lei, Xingyu Na, Mingbin Xu, Ernest Pusateri, Christophe Van Gyel, Yuanyuan Zhang, Shiyi Han (character ai), Zhen Huang
Delayed fusion: Integration of large language models in the first -step decoding in extreme to extreme voice recognition
Tear Hori, Martin Kocour (Brno Technological University), Adnan Haider, Erik McDermott, Xiaodan Zhuang
Exploring prediction objectives in masked prediction models for Speech Foundation
The living gods (Chinchen) Assember, educated, Zaldiel, are Alex.
Unbeat
Moji Heydari (University of Rochester), Mehrez Souden, Josh Atkins, Bruno Conejo
Speaker-ILP: Learning not supervised of speaker characteristics with pseudo-labels based on vector I
Shinji Watanabe (Carnegie Mellon University), Jeeweon Jung (Carnegie Mellon University), Ahmed Hussen Abdelaziz, Takuya Higuchi, Zak Aldeneh, Li-Wei Chen (Carnegie Mellon University), Stephen Shum, Tatian Likomomanenko, Barry-Jame Obry-Ju
Recovery correction of the voice recognition errors of entity appointed
Ernest Pusteri, Anirudh Kashi, Bortik Bandyopadhyay, Nadia Hyder, Sayanan Mahinder, Ravitaja Anantha, Daben Liu (Capital One), Sashank Gondala (more ai)
Selma: A speaking language model for virtual assistant interactions
Dominik Wagner
Slippery: exploiting the subsequent for the keyword stain with length limitation
Kumari Nishu, Minsik Cho, Devang Naik
Towards the automatic evaluation of self-supervised speech models using rank
Zadok, Lamai's vigma.
Expressions of gratitude
Tatiana Likhomanenko is an area chair and reviewer for ICASSP 2025.
Arnav Kundu, Aswin Sivaraman, Kunal Jain, Kuan-Lin Chen, Kumari Nishu, Nimish Venkat Marigo, Parnia Bahar, Sameer Badaskar, Takaki Hori, Tatiana Likomanko, Venki Nagesha and Zak Alden are reviewers for ICASSP 2025.
Venkataramanan Subramanian presents in the ICASSP 2025 industry forum.
