Search Results - "Porjazovski, Dejan"

From Raw Speech to Fixed Representations: A Comprehensive Evaluation of Speech Embedding Techniques by Porjazovski, Dejan, Grosz, Tamas, Kurimo, Mikko

“…Speech embeddings, fixed-size representations derived from raw audio data, play a crucial role in diverse machine learning applications. Despite the abundance…”
Get full text

Lahjoita puhetta: a large-scale corpus of spoken Finnish with some benchmarks by Moisio, Anssi, Porjazovski, Dejan, Rouhe, Aku, Getman, Yaroslav, Virkkunen, Anja, AlGhezi, Ragheb, Lennes, Mietta, Grósz, Tamás, Lindén, Krister, Kurimo, Mikko

“…The Donate Speech campaign has so far succeeded in gathering approximately 3600 h of ordinary, colloquial Finnish speech into the Lahjoita puhetta ( Donate…”
Get full text

Out-of-distribution generalisation in spoken language understanding by Porjazovski, Dejan, Moisio, Anssi, Kurimo, Mikko

“…Test data is said to be out-of-distribution (OOD) when it unexpectedly differs from the training data, a common challenge in real-world use cases of machine…”
Get full text

Topic Identification For Spontaneous Speech: Enriching Audio Features With Embedded Linguistic Information by Porjazovski, Dejan, Grósz, Tamás, Kurimo, Mikko

“…Traditional topic identification solutions from audio rely on an automatic speech recognition system (ASR) to produce transcripts used as input to a text-based…”
Get full text

Advancing Audio Emotion and Intent Recognition with Large Pre-Trained Models and Bayesian Inference by Porjazovski, Dejan, Getman, Yaroslav, Grósz, Tamás, Kurimo, Mikko

“…Large pre-trained models are essential in paralinguistic systems, demonstrating effectiveness in tasks like emotion recognition and stuttering detection. In…”
Get full text

Topic Identification for Spontaneous Speech: Enriching Audio Features with Embedded Linguistic Information by Porjazovski, Dejan, Grosz, Tamas, Kurimo, Mikko

“…Traditional topic identification solutions from audio rely on an automatic speech recognition system (ASR) to produce transcripts used as input to a text-based…”
Get full text

Lahjoita puhetta -- a large-scale corpus of spoken Finnish with some benchmarks by Moisio, Anssi, Porjazovski, Dejan, Rouhe, Aku, Getman, Yaroslav, Virkkunen, Anja, Grósz, Tamás, Lindén, Krister, Kurimo, Mikko

“…The Donate Speech campaign has so far succeeded in gathering approximately 3600 hours of ordinary, colloquial Finnish speech into the Lahjoita puhetta (Donate…”
Get full text