WAVLab | @CarnegieMellon

Shinji Watanabe's Audio and Voice Lab | WAVLab @LTIatCMU @SCSatCMU | Speech Recognition, Speech Enhancement, Spoken Language Understanding, and more.

[arXiv Sound @ArxivSound]

Masao, Someki, et al., "PlanRAG-Audio: Planning and Retrieval Augmented Generation for Long-form Audio Understanding,", https://t.co/rH3a96PWCN

[Shinji Watanabe @shinjiw_at_cmu]

We are looking for a postdoctoral researcher in speech and audio processing, with a possible start in the Fall 2026 semester. If you are interested in working with us, please apply through the following form: https://t.co/ssjinOn6LW

[William Chen @chenwanch1]

Accepted to ICML! See y’all in Korea 🇰🇷

[WAVLab | @CarnegieMellon @WavLab]

7. Phonological Tokenizer: Prosody-Aware Phonetic Token via Multi-Objective Fine-Tuning With Differentiable K-Means
Poster: May 6, 14:00
https://t.co/akYBgQc7v2

8. Online Register for Dual-Mode Self-Supervised Speech Models
Poster: May 7, 09:00
https://t.co/SgVx6G0orr

5/5

[WAVLab | @CarnegieMellon @WavLab]

WAVLab @ #ICASSP2026

We will present 8 papers at ICASSP in Barcelona. If you are attending, please stop by the talks/posters and chat with the authors.

arXiv links and presentation info below.

1/5

[WAVLab | @CarnegieMellon @WavLab]

5. Full-Duplex-Bench V1.5: Evaluating Overlap Handling for Full-Duplex Speech Models
Poster: May 8, 14:00
https://t.co/5487EFk8IQ

6. CALM: Joint Contextual Acoustic-Linguistic Modeling for Personalization of Multi-Speaker ASR
Oral: May 8, 15:00
https://t.co/6CdqYCBKXI

4/5

[WAVLab | @CarnegieMellon @WavLab]

3. Reasoning Beyond Majority Vote: An Explainable SpeechLM Framework for Speech Emotion Recognition
Oral: May 7, 15:00
https://t.co/kJnKYyGs1O

4. 2025 URGENT Speech Enhancement Challenge Multilingual P.808 Listening Tests
Oral: May 6, 17:50
https://t.co/ZlK6T5j7RV

3/5

[WAVLab | @CarnegieMellon @WavLab]

1. ICASSP 2026 URGENT Speech Enhancement Challenge
Poster: Fri May 8, 14:00 to 16:00, Poster Area 43
https://t.co/klUYslOO8e

2. SSVD-O: Parameter-Efficient Fine-Tuning with Structured SVD for Speech Recognition
Oral: Fri May 8, 10:00 to 10:20
https://t.co/gj2dy4keFN

2/5

[WAVLab | @CarnegieMellon @WavLab]

Congrats to Brian @brianyan918 on finishing his PhD defense today! It was great to see so many people show up for this big event and celebrate such an important milestone. Wishing you all the best in what comes next!

[Shinji Watanabe @shinjiw_at_cmu]

6 papers (4 main and 2 findings) were accepted at #ACL2026! All are speech papers :)

[arXiv Sound @ArxivSound]

Shikhar Bharadwaj, Chin-Jou Li, Kwanghee Choi, Eunjung Yeo, William Chen, Shinji Watanabe, David R. Mortensen, "An Empirical Recipe for Universal Phone Recognition," https://t.co/YRQ1a83nNs

[WAVLab | @CarnegieMellon @WavLab]

Congratulations to Li-Wei @liweiche77 on successfully defending his PhD today! 🎉 Wishing him all the best in his next chapter!

[WAVLab | @CarnegieMellon @WavLab]

Congratulations to Siddhant @Sid_Arora_18 on a successful PhD defense today! It was wonderful to celebrate this big milestone together. Wishing him all the best for the exciting journey ahead.

[Natural Language Processing Papers @HEI]

PRiSM: Benchmarking Phone Realization in Speech Models

Shikhar Bharadwaj, Chin-Jou Li, Yoonjae Kim, Kwanghee Choi, Eunjung Yeo, Ryan Soh-Eun Shim, Hanyu Zhou, Brendon Boldt, Karen Rosero Jacome, Kalvin Chang, Darsh Agrawal, …
https://t.co/wTZiVozMGA [𝚌𝚜.𝙲𝙻 𝚌𝚜.𝚂𝙳]

[arXiv Sound @ArxivSound]

Chenda Li, Wei Wang, Marvin Sach, Wangyou Zhang, Kohei Saijo, Samuele Cornell, Yihui Fu, Zhaoheng Ni, Tim Fingscheidt, Shinji Watanabe, Yanmin Qian, "ICASSP 2026 URGENT Speech Enhancement Challenge," https://t.co/Hupa3JJDKP

[arXiv Sound @ArxivSound]

Pu Wang, Shinji Watanabe, Hugo Van hamme, "SSVD-O: Parameter-Efficient Fine-Tuning with Structured SVD for Speech Recognition," https://t.co/vM5sgTuMAL

[arXiv Sound @ArxivSound]

Shih-Heng Wang, Jiatong Shi, Jinchuan Tian, Haibin Wu, Shinji Watanabe, "Do Neural Codecs Generalize? A Controlled Study Across Unseen Languages and Non-Speech Tasks," https://t.co/hCQOKgxCga

[jiatongshi @jiatongshi]

Heading to NeurIPS 2025 in San Diego!

I’ll present our spotlight poster, ARECHO, focusing on speech multi-metric estimation.

📍 Exhibit Hall C,D,E #2000
🗓️ Thu Dec 4, 11 a.m.–2 p.m. PST

If you’re around, let’s say hi or grab a coffee!

[jiatongshi @jiatongshi]

This is exactly the reason we worked for ESPnet-Codec, but being really hard to keep tracking as people are fast nowadays.

The similar issue happens at most speech tasks from ASR, TTS, to general speech LLM. It's a bit sad time for driving scientific findings 🥲

[jiatongshi @jiatongshi]

Speech isn’t just sound -> it’s how we turn thought into expression.

Our new work, Speech-DRAME, measures how well speech AI can act, aligning evaluation with human perception.

Paper: https://t.co/QzDuj2eJ6Z
Code: https://t.co/dOMqq8rgFJ