Speech AI · Speaker Embeddings · Similarity SearchBuilt an exploratory proof of concept
Speaker Similarity — Proof of Concept
A speech-AI proof of concept exploring speaker-embedding similarity to identify repeat callers in a telecom scenario.
Status
Exploratory proof of concept
The Problem
Identifying repeat callers efficiently in customer care workflows helps route inquiries and personalize interactions.
Solution & Contribution
Built a proof of concept using speaker embedding models to compute acoustic similarity across audio snippets and evaluate repeat-caller matching.
My Contribution
Audio processing pipeline, embedding extraction and similarity comparison logic.
Technologies Used
PythonSpeech EmbeddingsAudio ProcessingVector SimilaritySpeech AI
Interested in collaborating?
Get in touch with Khalid to discuss applied AI engineering, LLMs, or agent systems.
kh.regui@gmail.com