About
I define myself as a high-performing, intrinsically motivated individual with a strong passion for technology. I thrive in multidisciplinary
and multicultural teams, and I am motivated by projects with social impact, where technical challenges are combined with user needs.
Audio AI Engineer & Machine Learning Engineer.
- Blog: nahue-passano.github.io/blog
- City: Buenos Aires, Argentina
- Degree: Engineering
- Email: nahue.passano@gmail.com
🧠 I am an Audio AI and Machine Learning Engineer, currently focused on developing real-time AI systems for speech processing, including speaker separation, dereverberation, and denoising.
👁️ I have also gained experience in end-to-end computer vision projects and applied AI in the healthcare sector, particularly in analyzing and predicting the progression of ALS in patients.
🗣️ For my Acoustical Engineering thesis, I developed a voice cloning system for the Rioplatense Spanish dialect. This system synthesizes high-quality speech in real time, offering detailed control over pitch and phoneme duration.
📚 In addition, I have teaching experience in engineering and data science, where I contribute to skill development in these areas.
💡 I am passionate about leveraging technology to create innovative solutions that positively impact society.
Skills
Throughout my career, I have had the privilege of being a member of high-performance technical teams focused on various topics
within the realm of artificial intelligence. During this projects, I had the opportunity to develop and refine these skills, among others.
Resume
I've had the opportunity to collaborate with captivating interdisciplinary teams, enriching my experience profoundly.
Concurrently, I've engaged in compelling and groundbreaking projects that I take great pride in showcasing
Education
Acoustical Engineering
2016 - 2024
National University of Tres de Febrero, Buenos Aires, Argentina
Thesis: Development of an automatic audiobook system with Rioplatense prosody.
Bachelor’s + Master’s degree in Acoustics with a focus on advanced concepts in acoustic measurements for room control and sensing, electroacoustics for speaker design, audio electronics for the development of specialized equipment, and digital signal processing (DSP) for the development of specific audio and acoustics software.
Projects
AIRA: Ambisonics Impulse Response Analyzer
Innovative software for visualizing impulse responses from Ambisonics microphones, featuring an interactive interface that allows for detailed examination of sound reflections using ’hedgehog-type’ graphs
Mi identidad vocal
Development of voice cloning software for the former national deputy Jorge Rivas. Through his sessions in the Chamber of Deputies, his voice was successfully reconstructed, restoring his vocal identity.
Certificates
Cloud Computing (AWS)
CoderHouse
Convolutional Neural Networks
DeepLearning.ai
Data analysis with Python
freeCodeCamp
Docker for the absolute begginer
KodeKloud
Good Clinical Practices
CITI Program
MLOps concepts
DataCamp
Sequence Models
DeepLearning.ai
Software Engineering for Data Scientists in Python
DataCamp
Professional Experience
Sr Audio AI ENgineer
3/2024 - Present
Fusemachines
- R&D of speech denoising and dereverberation algorithms in real-time.
- Research into state-of-the-art literature on speech enhancement techniques.
- Development of datasets for speech separation, enhancement, and dereverberation.
- Design and deployment of web applications with Streamlit for model implementation.
Ssr Machine Learning Engineer
9/2023 - 3/2024
EverythingALS
- Presentation of scientific advancements to high-level executives, emphasizing skills in communication and specialized information management.
- Development of signal processing pipelines for the extraction of digital biomarkers from speech and measurements of pressure sensors, accelerometers and gyroscopes.
- Training of supervized models to predict subjective metrics of ALS patients.
- Design of a proof of concept on voice cloning for ALS patients using Eleven Labs, TortoiseTTS and RVC.
Ssr Machine Learning Engineer
4/2023 - 10/2023
MecanTronic
- Technical leadership in an Audio Machine Learning project (Voice Cloning for Argentinian speakers).
- Design technical exam for the selection process, perform technical interviews
- Machine Learning projects lifecycle managment with MLFlow.
- Implementation of CI pipelines with GitHub Actions for automated testing.
- Cloud architect with AWS for Machine Learning projects.
- Containerization of Machine Learning applications using Docker.
Jr Machine Learning Engineer
2/2022 - 4/2023
MecanTronic
- EDA for Object Detection and Instance Segmentation datasets.
- Reserch and application of Computer Vision models for object detection in drone imagery.
- Adaptation of libraries such as Detectron2 and Anomalib for instance segmentation and anomaly detection tasks in drone imagery.
- Full pipeline development of preprocessing, inference and postprocessing in Computer Vision projects.
- Training AI models on GPU cloud instances (AWS EC2, Paperspace).
Portfolio
During my academic and professional career, I have developed innovative and highly motivating projects. The most outstanding ones are listed below
- All
- Artificial Ingelligence
- Acoustics
- Digital Signal Processing
