{ "$schema": "https://raw.githubusercontent.com/jsonresume/resume-schema/v1.0.0/schema.json", "basics": { "name": "Gabriel Bibbó", "label": "Audio ML researcher and engineer", "image": "https://gbibbo.github.io/homepage_files/profile.jpg", "email": "gabobibbo@gmail.com", "url": "https://gbibbo.github.io/", "summary": "I build and evaluate machine-learning and audio-processing systems that have to work outside the notebook, from audio-language models and acoustic privacy to real-time audio software and tools for musicians.", "location": { "city": "Montevideo", "countryCode": "UY", "region": "Montevideo" }, "profiles": [ { "network": "LinkedIn", "username": "gabriel-bibbo", "url": "https://www.linkedin.com/in/gabriel-bibbo/" }, { "network": "GitHub", "username": "gbibbo", "url": "https://github.com/gbibbo" }, { "network": "Google Scholar", "username": "", "url": "https://scholar.google.com/citations?user=KEwHUaMAAAAJ&hl=es&oi=ao" }, { "network": "ORCID", "username": "0009-0003-2493-7412", "url": "https://orcid.org/0009-0003-2493-7412" } ] }, "work": [ { "name": "Edge Audio Labs", "position": "ML/DSP Engineer", "url": "https://edgeaudiolabs.com/", "location": "Montevideo, Uruguay", "startDate": "2026-06", "endDate": "", "summary": "ML/DSP product work spanning neural audio rendering, real-time detection, measurement, and perceptual evaluation.", "highlights": [ "Applied machine learning, digital signal processing, testing, and perceptual evaluation across two confidential audio product lines, without disclosing client or project identities.", "Designed and delivered a rendering-side feature that maps score dynamics to model-level timbral expression rather than post-render gain alone, after reverse-engineering the end-to-end audio pipeline and identifying a hidden control-path failure.", "Built measurement and listening-test tooling covering approximately 580 renders and a 48-clip blind evaluation, then delivered the feature server-side without retraining the model.", "Built a headless C++ evaluation pipeline for real-time note and onset detection, from WAV and MIDI inputs through the production DSP to JSON metrics, regression tests, and adversarial canary cases.", "Found and corrected a systematic onset timing offset of approximately 104 ms, improved detector guard logic, and communicated results through technical documentation, pull requests, Jira, and client-facing presentations." ] }, { "name": "University of Surrey", "position": "Visiting Researcher (collaboration)", "url": "https://www.surrey.ac.uk/", "location": "Remote", "startDate": "2025-12", "endDate": "", "summary": "Ongoing research collaboration on robust VAD and privacy-preserving machine listening.", "highlights": [ "Preparing the manuscript ‘A Psychometric Evaluation of Audio-Language Models for Robust Voice Activity Detection’ for Elsevier Computer Speech & Language with Mark D. Plumbley and Simone Spagnol.", "Co-authoring work with Arshdeep Singh and Mark D. Plumbley on privacy-preserving audio and machine listening." ] }, { "name": "University of Surrey", "position": "Research Engineer in Sound Sensing", "url": "https://www.surrey.ac.uk/", "location": "Guildford, UK", "startDate": "2022-11", "endDate": "2025-11", "summary": "Research engineering for real-world sound sensing, privacy-preserving audio, and edge deployment.", "highlights": [ "Developed end-to-end audio ML systems for real-world smart environments, covering data preparation, model evaluation, prototype deployment, open-source releases, demos, datasets, and technical documentation.", "Built privacy-preserving SED pipelines for sensitive in-home recordings, including a 197 GB residential audio dataset, speech-removal workflows, and reproducible evaluation resources.", "Built an eight-model VAD benchmark on CHiME-Home and, separately, evaluated audio-language models under controlled duration, noise, reverberation, and spectral degradations.", "Deployed real-time CNN inference on Raspberry Pi, including latency, thermal, efficiency, and robustness evaluation for edge sound sensing.", "Published and presented research at ICASSP, IEEE WASPAA, CHiME Workshop, Inter-Noise, SMC, UKAI, UKIS, and AES. Supervised undergraduate and master’s projects." ] }, { "name": "Webhelp", "position": "Technical Support Engineer - Google Workspace", "url": "https://workspace.google.com/", "location": "Barcelona, Spain", "startDate": "2022-03", "endDate": "2022-11", "summary": "Enterprise Google Workspace support across identity, APIs, migration, and configuration.", "highlights": [ "Tier 3 support for Google Workspace enterprise customers across APIs, OAuth, SAML/SSO, IAM, user provisioning, data migration, DNS/domain configuration, and security/compliance settings." ] }, { "name": "KPMG", "position": "IT Auditor", "url": "https://kpmg.com/es/es.html", "location": "Barcelona, Spain", "startDate": "2021-11", "endDate": "2022-03", "summary": "Technology audit work for telecommunications companies and IT organisations.", "highlights": [ "Supported telecommunications companies and IT departments in technology audit engagements." ] }, { "name": "Ikatu", "position": "R&D Engineer", "url": "https://www.ikatu.com/", "location": "Montevideo, Uruguay", "startDate": "2016-08", "endDate": "2019-12", "summary": "Embedded C/C++ audio and home-automation engineering for product-facing Bang & Olufsen systems.", "highlights": [ "Designed and shipped embedded C/C++ audio and IoT firmware for Bang & Olufsen home automation products, including low-level drivers, hardware integration, audio I/O, and Internet connectivity.", "Worked across requirements, architecture, implementation, testing, validation, and customer-facing documentation.", "Trained and onboarded incoming programmers in embedded development practices." ] }, { "name": "Ikatu", "position": "Engineering Intern", "url": "https://www.ikatu.com/", "location": "Montevideo, Uruguay", "startDate": "2016-04", "endDate": "2016-07", "summary": "Home-automation project work across engineering implementation and coordination.", "highlights": [ "Developed and coordinated a complete home automation system project before transitioning into the R&D Engineer role." ] } ], "education": [ { "institution": "Universitat Pompeu Fabra", "area": "Computational models, audio engineering, perception, cognition, and interactive systems", "studyType": "MSc Sound and Music Computing", "startDate": "2020-09", "endDate": "2021-08", "score": "", "courses": [ "60 ECTS", "Master thesis on harmonic compatibility for EDM mixing. Final thesis grade: 9/10." ] }, { "institution": "Universidad de la República - Facultad de Ingeniería", "area": "Specialisation in electronics and signal processing", "studyType": "BSc Electrical Engineering", "startDate": "2012-02", "endDate": "2017-12", "score": "", "courses": [ "Equivalent to 300 ECTS", "Bachelor thesis on autonomous mobile robots communicated by software-defined radio." ] }, { "institution": "School Nº265 ‘Virgilio Scarabelli Alberti’", "area": "Musical language, choral singing, guitar, and instrumental ensembles", "studyType": "Formal musical training", "startDate": "2002-03", "endDate": "2005-12", "score": "", "courses": [] } ], "publications": [ { "name": "Privacy for Audio AI: Risks, Challenges, and Emerging Solutions in the Era of Audio AI [Panel discussion]", "publisher": "2025 AES International Conference on Artificial Intelligence and Machine Learning for Audio", "releaseDate": "2025", "url": "https://openresearch.surrey.ac.uk/esploro/outputs/991036566602346", "summary": "Authors: Thomas Deacon, Jennifer Williams, Jason R. C. Nurse, Christopher Hicks, Gabriel Bibbó, Arshdeep Singh, Mark D. Plumbley" }, { "name": "Speech Removal Framework for Privacy-preserving Audio Recordings", "publisher": "2025 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), Tahoe City, CA, October 2025", "releaseDate": "2025", "url": "https://zenodo.org/records/17050321", "summary": "Authors: Gabriel Bibbó, Arshdeep Singh, Thomas Deacon, Mark D. Plumbley" }, { "name": "Room Acoustics and Microphone Characteristics Show Systematic Impact on Sound Event Recognition", "publisher": "Proceedings of the 54th International Congress and Exposition on Noise Control Engineering, São Paulo, Brazil, August 2025", "releaseDate": "2025", "url": "https://www.even3.com.br/anais/international-congress-exposition-noise-control-engineering/1070751-room-acoustics-and-microphone-characteristics-show-systematic-impact-on-sound-event-recognition/", "summary": "Authors: Gabriel Bibbó, Craig Cieciura, Mark D. Plumbley" }, { "name": "Integrating IP broadcasting with audio tags: Workflow and challenges", "publisher": "2025 AES International Conference on Artificial Intelligence and Machine Learning for Audio", "releaseDate": "2025", "url": "https://openresearch.surrey.ac.uk/esploro/outputs/conferencePaper/Integrating-IP-broadcasting-with-audio-tags/991013165302346?institution=44SUR_INST", "summary": "Authors: Rhys Burchett-Vass, Arshdeep Singh, Gabriel Bibbó, Mark D. Plumbley" }, { "name": "Soundscape Experience Mapping: A Deep Listening Approach for Eliciting Older Adults' Perceptions of Indoor Soundscapes", "publisher": "Forum Acusticum / Euronoise 2025, Málaga, Spain, June 2025", "releaseDate": "2025", "url": "https://openresearch.surrey.ac.uk/esploro/outputs/conferencePresentation/Soundscape-experience-mapping-A-deep-listening/99994066602346?institution=44SUR_INST", "summary": "Authors: Thomas Deacon, Gabriel Bibbó, Arshdeep Singh, Mark D. Plumbley" }, { "name": "Personalized Live Sound Recognition Using Efficient PANNs [Show and Tell]", "publisher": "IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025), Hyderabad, India, April 2025", "releaseDate": "2025", "url": "https://zenodo.org/records/15498188", "summary": "Authors: Arshdeep Singh, Haohe Liu, Gabriel Bibbó, Thomas Deacon, Mark D. Plumbley" }, { "name": "Environmental sound classification on an embedded hardware platform", "publisher": "INTER-NOISE and NOISE-CON Congress and Conference Proceedings, Nantes, France, August 2024", "releaseDate": "2024", "url": "http://dx.doi.org/10.3397/in_2024_3723", "summary": "Authors: Gabriel Bibbó, Arshdeep Singh, Mark D. Plumbley" }, { "name": "The Sounds of Home: A Speech-Removed Residential Audio Dataset for Sound Event Detection", "publisher": "8th International Workshop on Speech Processing in Everyday Environments (CHiME 2024), Kos Island, Greece, September 2024", "releaseDate": "2024", "url": "http://dx.doi.org/10.21437/chime.2024-11", "summary": "Authors: Gabriel Bibbó, Thomas Deacon, Arshdeep Singh, Mark D. Plumbley" }, { "name": "Soundscape Personalisation at Work: Designing AI-Enabled Sound Technologies for the Workplace", "publisher": "International Conference on Sound and Music Computing (SMC 2024), Porto, Portugal, July 2024", "releaseDate": "2024", "url": "https://smcnetwork.org/smc2024/papers/SMC2024_paper_id117.pdf", "summary": "Authors: Thomas Deacon, Gabriel Bibbó, Arshdeep Singh, Mark D. Plumbley" }, { "name": "Recognise and Notify Sound Events Using a Raspberry PI Based Standalone Device [Demo]", "publisher": "IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA 2023), New York, U.S.A, October 2023", "releaseDate": "2023", "url": "http://dx.doi.org/10.5281/zenodo.15465882", "summary": "Authors: Gabriel Bibbó, Arshdeep Singh, Mark D. Plumbley" }, { "name": "A New Compatibility Measure for Harmonic EDM Mixing", "publisher": "International Conference on Web Engineering (ICWE 2022), Bari, Italy, July 2022", "releaseDate": "2022", "url": "http://dx.doi.org/10.1007/978-3-031-09917-5_37", "summary": "Authors: Gabriel Bibbó, Ángel Faraldo" }, { "name": "Towards a New Compatibility Measure for Harmonic EDM Mixing", "publisher": "Master thesis, Universitat Pompeu Fabra, Barcelona, Spain, 2021. Supervisor: Ángel Faraldo.", "releaseDate": "2021", "url": "http://dx.doi.org/10.5281/zenodo.5554688", "summary": "Authors: Gabriel Bibbó" }, { "name": "Autonomous Mobile Robots Communicated by Software Defined Radio", "publisher": "Bachelor thesis, Universidad de la República, Montevideo, Uruguay, 2017. Supervisors: Pablo Belzarena and Federico Larroca.", "releaseDate": "2017", "url": "https://iie.fing.edu.uy/publicaciones/2017/BGR17/", "summary": "Authors: Gabriel Bibbó, Mariana Gelós, Martín Randall" } ], "skills": [ { "name": "Stack", "level": "", "keywords": [ "Python", "C/C++", "PyTorch", "Hugging Face", "PEFT", "TorchAudio", "librosa", "Essentia", "mido", "scikit-learn", "pandas", "NumPy", "SciPy", "Flask", "FastAPI", "Streamlit", "Docker", "Git", "Linux CLI", "Bash", "Slurm", "Redis", "Prometheus", "Grafana", "PostgreSQL", "SQLite", "MATLAB", "Unreal Engine 5.4", "FMOD", "VS Code" ] }, { "name": "ML", "level": "", "keywords": [ "CNNs", "Transformers", "Audio-Language Models", "LoRA Fine-tuning", "4-bit Quantization", "Supervised and Self-supervised Learning", "Evaluation Pipelines", "Statistical Testing", "Edge Deployment" ] }, { "name": "Audio", "level": "", "keywords": [ "Sound Event Detection", "Voice Activity Detection", "Pitch and Onset Detection", "Music Information Retrieval", "Digital Signal Processing", "Real-Time Audio", "Perceptual Evaluation", "DAWs", "Ableton", "DJing", "Electronic Music Production" ] }, { "name": "Practice", "level": "", "keywords": [ "Reproducible ML pipelines", "Automated Audio Testing", "Dataset Curation", "Open-Source Development", "MLOps practices", "AI-assisted Development", "Technical Writing", "Interdisciplinary Collaboration" ] } ], "languages": [ { "language": "Spanish", "fluency": "Native" }, { "language": "English", "fluency": "C1 (CEFR)" }, { "language": "Portuguese", "fluency": "A2 (CEFR)" } ], "projects": [ { "name": "Audio-Language Models for Voice Activity Detection", "description": "This research evaluates how audio-language models detect speech when the audio is short, noisy, reverberant, or filtered. It compares Qwen2-Audio-7B, Qwen2-Audio-7B with LoRA, Qwen3-Omni-30B, and Silero VAD. The best result came from Qwen2-Audio-7B with LoRA and OPRO-Template: 93.3% balanced accuracy on 21,340 degraded clips.", "highlights": [ "Audio AI · VAD · Model evaluation" ], "keywords": [ "Qwen", "LoRA", "OPRO", "Silero", "PyTorch" ], "url": "https://github.com/gbibbo/qwen-vad-lora", "roles": [], "entity": "", "type": "application" }, { "name": "Recovery Gain in Pruned Text-to-Audio Diffusion", "description": "Recovery fine-tuning can make a heavily pruned text-to-audio model look largely restored when it is evaluated under conditions similar to those used for fine-tuning. But that apparent recovery weakens sharply for shorter clips or out-of-domain prompts. The result shows why compressed generative models should be evaluated across real operating conditions, not at a single benchmark setting.", "highlights": [ "Text-to-audio · Model compression · Evaluation" ], "keywords": [ "AudioLDM", "Diffusion", "Structured pruning", "CLAP", "PyTorch" ], "url": "https://github.com/gbibbo/audioldm-modality-swap-pruning", "roles": [], "entity": "", "type": "application" }, { "name": "ASR Enhancement Platform", "description": "An end-to-end MVP for comparing raw transcription with enhance-and-transcribe on pre-recorded audio. The backend persists jobs, audio artifacts, transcripts, and provider payloads, with FastAPI, Celery, PostgreSQL, Redis, MinIO, Docker Compose, metrics, tracing, Grafana, and CI. It is a reproducible engineering prototype, not a production-hardened service.", "highlights": [ "Speech enhancement · Backend platform" ], "keywords": [ "FastAPI", "Celery", "Docker", "PostgreSQL", "Redis" ], "url": "https://github.com/gbibbo/asr_enhancement", "roles": [], "entity": "", "type": "application" }, { "name": "Traktor ML", "description": "Traktor ML turns a local Techno and Tech House library into Traktor-ready playlists. The pipeline extracts MERT embeddings, separates stems with Demucs, reads BPM and key metadata with Essentia, clusters similar tracks, orders them for smoother transitions, and exports M3U playlists. The current V4 run processed 239 tracks and exported 14 playlists.", "highlights": [ "MIR · DJ library organisation" ], "keywords": [ "MERT", "Demucs", "Essentia", "HDBSCAN", "UMAP", "Streamlit" ], "url": "https://github.com/gbibbo/traktor", "roles": [], "entity": "", "type": "application" }, { "name": "Speech Removal Framework", "description": "A framework for removing speech from audio recordings before they are shared or published. The system supports privacy-preserving release workflows while retaining non-speech acoustic information for sound event detection research.", "highlights": [ "Privacy · Speech removal · WASPAA" ], "keywords": [ "PANNs", "AST", "Silero VAD", "WebRTC VAD", "Audio privacy" ], "url": "https://huggingface.co/spaces/gbibbo/vad_demo", "roles": [], "entity": "", "type": "application" }, { "name": "Sounds of Home Dataset", "description": "Sounds of Home is a residential audio dataset for sound event detection. It contains 1,344 one-hour recordings collected from 8 participants in Belgium, using AudioMoth recorders placed in living rooms and kitchens. Speech was removed before release, and PANNs predictions were provided for the audio frames.", "highlights": [ "Privacy-preserving dataset · Domestic audio" ], "keywords": [ "SED", "AudioMoth", "PANNs", "Datasets" ], "url": "https://www.cvssp.org/data/ai4s/sounds_of_home/", "roles": [], "entity": "", "type": "application" }, { "name": "Raspberry Pi Sound Event Recognition Demo", "description": "Raspberry Pi demo for real-time sound event recognition. The system runs pre-trained neural networks on a low-cost edge device, exposes a web interface, and can send email notifications when selected AudioSet events are detected.", "highlights": [ "Edge AI · Sound event recognition" ], "keywords": [ "Raspberry Pi", "AudioSet", "Real-time inference", "Email notifications" ], "url": "https://github.com/gbibbo/pisoundsensing", "roles": [], "entity": "", "type": "application" }, { "name": "Harmonic EDM Mixing Compatibility", "description": "This music analysis system estimates how well two EDM tracks mix harmonically. It analyzes tracks, computes chroma features, converts them into Tonal Interval Vectors, compares harmonic compatibility, and suggests pitch shifts that can improve a mix. The work began as an MSc thesis and later became an ICWE 2022 publication.", "highlights": [ "MIR · Harmonic mixing · MSc thesis" ], "keywords": [ "Chroma", "TIV", "Essentia", "librosa", "EDM" ], "url": "https://github.com/gbibbo/harmonic_mix", "roles": [], "entity": "", "type": "application" }, { "name": "3H-ATO", "description": "Mechanical tool designed during the pandemic to avoid touching shared surfaces directly.", "highlights": [ "Product design · Mechanical prototyping" ], "keywords": [ "Mechanical design", "Prototyping", "Product development" ], "url": "https://www.youtube.com/watch?v=aQVo0i5OLWU", "roles": [], "entity": "", "type": "application" }, { "name": "Automatic IoT Soap Dispenser", "description": "IoT handwashing device for industrial environments. The device used stainless steel, WiFi, cloud connectivity, IR/RFID sensors, and a 3-litre tank.", "highlights": [ "IoT · Embedded systems" ], "keywords": [ "WiFi", "IR/RFID", "Cloud connectivity", "Industrial hygiene" ], "url": "", "roles": [], "entity": "", "type": "application" }, { "name": "UyVoy Mobile App", "description": "Mobile app project for booking appointments and reducing crowding during the pandemic. Gabriel worked as product owner and project lead.", "highlights": [ "Product · Mobile app" ], "keywords": [ "Product ownership", "Civic tech", "Project management" ], "url": "", "roles": [], "entity": "", "type": "application" }, { "name": "ALPACA", "description": "Software engineering prototype for algorithmic trading infrastructure, with market data ingestion, event processing, risk controls, historical simulation, persistence, API access, and monitoring. It has not been tested in production.", "highlights": [ "Software engineering · Trading infrastructure" ], "keywords": [ "Python", "Backtesting", "Risk controls", "Monitoring" ], "url": "https://github.com/gbibbo/alpaca", "roles": [], "entity": "", "type": "application" } ], "meta": { "canonical": "https://gbibbo.github.io/", "version": "1.0.0", "lastModified": "2026-09-06" } }