Vasista Sai Lodagala

CS-FLEURS: A Massively Multilingual and Code-Switched Speech Dataset

Brian Yan, Injy Hamed, Shuichiro Shimizu, Vasista Sai Lodagala et autres

We present CS-FLEURS, a new dataset for developing and evaluating code-switched speech recognition and translation systems beyond high-resourced languages. CS-FLEURS consists of 4 test sets which cover in total 113 unique code-switched language pairs across 52 languages: 1) a 14 X-English language …

SawtArabi: A Benchmark Corpus for Arabic TTS. Standard, Dialectal and Code-Switching

Vasista Sai Lodagala, Lamya Alkanhal, Daniel Izham, Shivam Mehta et autres

CS-FLEURS: A Massively Multilingual and Code-Switched Speech Dataset

Brian Yan, Injy Hamed, Shuichiro Shimizu, Vasista Sai Lodagala et autres

Octopus: Towards Building the Arabic Speech LLM Suite

Sara Althubaiti, Vasista Sai Lodagala, Yousseif Ahmed Elshahawy, Daniel Izham et autres

Sara Althubaiti, Vasista Sai Lodagala, Tjad Clark, Yousseif Ahmed Elshahawy, Daniel Izham, Abdullah Alrajeh, Aljawahrah Bin Tamran, Ahmed Ali. Proceedings of The Third Arabic Natural Language Processing Conference. 2025.

All Ears: Building Self-Supervised Learning based ASR models for Indian Languages at scale

Vasista Sai Lodagala, Abhishek Ghose Biswas, Shoutrik Das, Frank Jordan et autres

SALTTS: Leveraging Self-Supervised Speech Representations for improved Text-to-Speech Synthesis

R Sivaguru, Vasista Sai Lodagala, S Umesh

SALTTS: Leveraging Self-Supervised Speech Representations for improved Text-to-Speech Synthesis

R Sivaguru, Vasista Sai Lodagala, Srinivasan Umesh

While FastSpeech2 aims to integrate aspects of speech such as pitch, energy, and duration as conditional inputs, it still leaves scope for richer representations. As a part of this work, we leverage representations from various Self-Supervised Learning (SSL) models to enhance the …

Data2vec-Aqc: Search for the Right Teaching Assistant in the Teacher-Student Training Setup

Vasista Sai Lodagala, Sreyan Ghosh, Srinivasan Umesh

In this paper, we propose a new Self-Supervised Learning (SSL) algorithm called data2vec-aqc, for speech representation learning from unlabeled speech data. Our goal is to improve SSL for speech in domains where both unlabeled and labeled data are limited. Building on the …

in, us (code pays fourni par la source)

CCC-WAV2VEC 2.0: Clustering AIDED Cross Contrastive Self-Supervised Learning of Speech Representations

Vasista Sai Lodagala, Sreyan Ghosh, Srinivasan Umesh

While Self-Supervised Learning has helped reap the benefit of the scale from the available unlabeled data, the learning paradigms are continously being bettered. We present a new pre-training strategy named ccc-wav2vec 2.0, which uses clustering and an augmentation based cross-contrastive loss as …

in, us (code pays fourni par la source)

PADA: Pruning Assisted Domain Adaptation for Self-Supervised Speech Representations

Vasista Sai Lodagala, Sreyan Ghosh, Srinivasan Umesh

While self-supervised speech representation learning (SSL) models serve a variety of downstream tasks, these models have been observed to overfit to the domain from which the unlabeled data originates. To alleviate this issue, we propose PADA (Pruning Assisted Domain Adaptation). Before performing …

in, us (code pays fourni par la source)

data2vec-aqc: Search for the right Teaching Assistant in the Teacher-Student training setup

Vasista Sai Lodagala, Sreyan Ghosh, S. Umesh

In this paper, we propose a new Self-Supervised Learning (SSL) algorithm called data2vec-aqc, for speech representation learning from unlabeled speech data. Our goal is to improve SSL for speech in domains where both unlabeled and labeled data are limited. Building on the …

Technology Pipeline for Large Scale Cross-Lingual Dubbing of Lecture Videos into Multiple Indian Languages

Anusha Prakash, Arun Kumar, Ashish Seth, Bhagyashree Mukherjee et autres

Cross-lingual dubbing of lecture videos requires the transcription of the original audio, correction and removal of disfluencies, domain term discovery, text-to-text translation into the target language, chunking of text using target language rhythm, text-to-speech synthesis followed by isochronous lipsyncing to the original …