Hearing voices at the National Library -- a speech corpus and acoustic model for the Swedish language
This paper explains our work in developing new acoustic models for automated speech recognition (ASR) at KBLab, the infrastructure for data-driven research at the National Library of Sweden (KB). We evaluate different approaches for a viable speech-to-text pipeline for audiovisual resources in Swedish, using the wav2vec 2.0 architecture in combination with speech corpuses created from KB's collections. These approaches include pretraining an acoustic model for Swedish from the ground up, and fine-tuning existing monolingual and multilingual models. The collections-based corpuses we use have been sampled from millions of hours of speech, with a conscious attempt to balance regional dialects to produce a more representative, and thus more democratic, model. The acoustic model this enabled, "VoxRex", outperforms existing models for Swedish ASR. We also evaluate combining this model with various pretrained language models, which further enhanced performance. We conclude by highlighting the potential of such technology for cultural heritage institutions with vast collections of previously unlabelled audiovisual data. Our models are released for further exploration and research here: https://huggingface.co/KBLab.
Code (0)
등록된 구현이 없습니다.
Tasks
speech-recognitionSpeech RecognitionSpeech-to-TextSimilar Papers 제목 키워드 기반
Dataset of British English speech recordings for psychoacoustics and speech processing research
The Clarity Speech Corpus is a forty speaker British English speech dataset. The corpus was created for the purpose of running listening tests to gauge speech intelligibility and quality in the Clarity Project, which has…
SentenceAudiobook Dialogues as Training Data for Conversational Style Synthetic Voices
Synthetic voices are increasingly used in applications that require a conversational speaking style, raising the question as to which type of training data yields the most suitable speaking style for such applications. T…
Sentencetext-to-speechText to SpeechSpeech Separation for Hearing-Impaired Children in the Classroom
Classroom environments are particularly challenging for children with hearing impairments, where background noise, multiple talkers, and reverberation degrade speech perception. These difficulties are greater for childre…
Transfer LearningSpeech SeparationEvaluating the Intelligibility Benefits of Neural Speech Enrichment for Listeners with Normal Hearing and Hearing Impairment using the Greek Harvard Corpus
In this work we evaluate a neural based speech intelligibility booster based on spectral shaping and dynamic range compression (SSDRC), referred to as WaveNet-based SSDRC (wSSDRC), using a recently designed Greek Harvard…
Speech EnhancementArVoice: A Multi-Speaker Dataset for Arabic Speech Synthesis
We introduce ArVoice, a multi-speaker Modern Standard Arabic (MSA) speech corpus with diacritized transcriptions, intended for multi-speaker speech synthesis, and can be useful for other tasks such as speech-based diacri…
DeepFake DetectionFace SwappingSpeech SynthesisVoice Conversion