8 papers
FlowW2N: Whispered-to-Normal Speech Conversion via Flow-Matching
Fabian Ritter-Gutierrez, Md Asif Jalal, Pablo Peso Parada +5
Whispered-to-normal (W2N) speech conversion aims to reconstruct missing phonation from whispered input while preserving content and speaker identity. This task is challenging due t…
Retrieval Augmented Generation based context discovery for ASR
Dimitrios Siskos, Stavros Papadopoulos, Pablo Peso Parada +3
This work investigates retrieval augmented generation as an efficient strategy for automatic context discovery in context-aware Automatic Speech Recognition (ASR) system, in order…
ValSub: Subsampling Validation Data to Mitigate Forgetting during ASR Personalization
Haaris Mehmood, Karthikeyan Saravanan, Pablo Peso Parada +5
Automatic Speech Recognition (ASR) is widely used within consumer devices such as mobile phones. Recently, personalization or on-device model fine-tuning has shown that adaptation…
Diffusion based Text-to-Music Generation with Global and Local Text based Conditioning
Jisi Zhang, Pablo Peso Parada, Md Asif Jalal +1
Diffusion based Text-To-Music (TTM) models generate music corresponding to text descriptions. Typically UNet based diffusion models condition on text embeddings generated from a pr…
persoDA: Personalized Data Augmentation for Personalized ASR
Pablo Peso Parada, Spyros Fontalis, Md Asif Jalal +6
Data augmentation (DA) is ubiquitously used in training of Automatic Speech Recognition (ASR) models. DA offers increased data variability, robustness and generalization against di…
Exploring compressibility of transformer based text-to-music (TTM) models
Vasileios Moschopoulos, Thanasis Kotsiopoulos, Pablo Peso Parada +7
State-of-the art Text-To-Music (TTM) generative AI models are large and require desktop or server class compute, making them infeasible for deployment on mobile phones. This paper…
Locality enhanced dynamic biasing and sampling strategies for contextual ASR
Md Asif Jalal, Pablo Peso Parada, George Pavlidis +8
Automatic Speech Recognition (ASR) still face challenges when recognizing time-variant rare-phrases. Contextual biasing (CB) modules bias ASR model towards such contextually-releva…
Consistency Based Unsupervised Self-training For ASR Personalisation
Jisi Zhang, Vandana Rajan, Haaris Mehmood +7
On-device Automatic Speech Recognition (ASR) models trained on speech data of a large population might underperform for individuals unseen during training. This is due to a domain…