Alex N Gichamba

I am a research associate at Carnegie Mellon University Africa working on speech technology. My work includes: multilingual speech recognition and speech translation. I am also working on zero-shot TTS. Much of this work happens through the WakandaAI project which is supported by Afretec, where we build open models and systems for under-resourced African languages.

I (sometimes) write explainer blogs of my papers or those I find interesting. You can check those out here technical posts.

Outside research, I love endurance sports and the process behind pushing physical limits. I document aspects of my training in my sport blog.

news

Jun 04, 2026	Probing Low Frame Rate Degradation in Neural Audio Codecs has been accepted at INTERSPEECH 2026. 🎉
Dec 09, 2024	Presented ColBERT Retrieval and Ensemble Response Scoring for LLM Question Answering at the IEEE GLOBECOM 2024 workshops in Cape Town.

latest posts

Jun 26, 2026	My First Marathon
Jun 26, 2026	[Explainer] Probing Low Frame Rate Degration in Neural Audio Codecs

selected publications

Interspeech

Probing Low Frame Rate Degradation in Neural Audio Codecs

Alex Gichamba and Moise Busogi

In Interspeech, 2026

arXiv Bib

@inproceedings{gichamba2026codec,
  author = {Gichamba, Alex and Busogi, Moise},
  title = {Probing Low Frame Rate Degradation in Neural Audio Codecs},
  booktitle = {Interspeech},
  publisher = {ISCA},
  year = {2026},
}

IJCNLP-AACL

AfriSpeech-MultiBench: A Verticalized Multidomain Multicountry Benchmark Suite for African Accented English ASR

Gabrial Zencha Ashungafac, Mardhiyah Sanni, Busayo Awobade, and 2 more authors

In IJCNLP-AACL, 2025

DOI arXiv Bib

@inproceedings{ashungafac2025afrispeech,
  author = {Ashungafac, Gabrial Zencha and Sanni, Mardhiyah and Awobade, Busayo and Gichamba, Alex and Olatunji, Tobi},
  title = {AfriSpeech-MultiBench: A Verticalized Multidomain Multicountry Benchmark Suite for African Accented English ASR},
  booktitle = {IJCNLP-AACL},
  pages = {3642--3653},
  publisher = {The Asian Federation of Natural Language Processing and The Association for Computational Linguistics},
  year = {2025},
  url = {https://doi.org/10.18653/v1/2025.ijcnlp-long.190},
  doi = {10.18653/V1/2025.IJCNLP-LONG.190},
}

GLOBECOM WKSHP

ColBERT Retrieval and Ensemble Response Scoring for Language Model Question Answering

Alex Gichamba, Tewodros Kederalah Idris, Brian Ebiyau, and 2 more authors

In IEEE GLOBECOM Workshops, 2024

DOI arXiv Bib

@inproceedings{gichamba2024colbert,
  author = {Gichamba, Alex and Idris, Tewodros Kederalah and Ebiyau, Brian and Nyberg, Eric and Mitamura, Teruko},
  title = {ColBERT Retrieval and Ensemble Response Scoring for Language Model Question Answering},
  booktitle = {IEEE GLOBECOM Workshops},
  pages = {1--7},
  publisher = {IEEE},
  year = {2024},
  url = {https://doi.org/10.1109/GCWkshp64532.2024.11100297},
  doi = {10.1109/GCWKSHP64532.2024.11100297},
}

Interspeech

ESPnet-SPK: Full Pipeline Speaker Embedding Toolkit with Reproducible Recipes, Self-Supervised Front-Ends, and Off-the-Shelf Models

Jee-weon Jung, Wangyou Zhang, Jiatong Shi, and 6 more authors

In Interspeech, 2024

DOI Bib

@inproceedings{jung2024espnetspk,
  author = {Jung, Jee-weon and Zhang, Wangyou and Shi, Jiatong and Aldeneh, Zakaria and Higuchi, Takuya and Gichamba, Alex and Theobald, Barry-John and Hussen Abdelaziz, Ahmed and Watanabe, Shinji},
  title = {ESPnet-SPK: Full Pipeline Speaker Embedding Toolkit with Reproducible Recipes, Self-Supervised Front-Ends, and Off-the-Shelf Models},
  booktitle = {Interspeech},
  publisher = {ISCA},
  year = {2024},
  url = {https://doi.org/10.21437/Interspeech.2024-1345},
  doi = {10.21437/INTERSPEECH.2024-1345}
}