Augmenting Polish Automatic Speech Recognition System With Synthetic Data
This paper presents a system developed for submission to Poleval 2024, Task 3: Polish Automatic Speech Recognition Challenge. We describe Voicebox-based speech synthesis pipeline and utilize it to augment Conformer and Whisper speech recognition models with synthetic data. We show that addition of synthetic speech to training improves achieved results significantly. We also present final results achieved by our models in the competition.
Code (0)
등록된 구현이 없습니다.
Tasks
Automatic Speech Recognitionspeech-recognitionSpeech RecognitionSpeech SynthesisSimilar Papers 제목 키워드 기반
Enhancements in statistical spoken language translation by de-normalization of ASR results
Spoken language translation (SLT) has become very important in an increasingly globalized world. Machine translation (MT) for automatic speech recognition (ASR) systems is a major challenge of great interest. This resear…
Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Machine TranslationSegmentation+5SpokesBiz -- an Open Corpus of Conversational Polish
This paper announces the early release of SpokesBiz, a freely available corpus of conversational Polish developed within the CLARIN-BIZ project and comprising over 650 hours of recordings. The transcribed recordings have…
Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech RecognitionFramework for Curating Speech Datasets and Evaluating ASR Systems: A Case Study for Polish
Speech datasets available in the public domain are often underutilized because of challenges in discoverability and interoperability. A comprehensive framework has been designed to survey, catalog, and curate available s…
Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech RecognitionEuronews: a multilingual speech corpus for ASR
In this paper we present a multilingual speech corpus, designed for Automatic Speech Recognition (ASR) purposes. Data come from the portal Euronews and were acquired both from the Web and from TV. The corpus includes dat…
Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Machine Translationspeech-recognition+1Polish Read Speech Corpus for Speech Tools and Services
This paper describes the speech processing activities conducted at the Polish consortium of the CLARIN project. The purpose of this segment of the project was to develop specific tools that would allow for automatic and …
Action DetectionActivity DetectionGrapheme-to-Phoneme ConversionKeyword Spotting+3