Project AURIONAI Use-Cases Research & Innovation
JAIST

JAIST

University, Japan. 3 models in the directory.

universityJapan
0deployments

Identity

Type
university
Country label
Japan
HQ country
Japan
Founding country
not in the notes
Parent
not in the notes
Parent country
not in the notes
Country basis
the research notes
Where this comes from
Convention D1: a dual label, HQ first, when the HQ, founding or parent country differ.

Models by Class

Models

ModelClassOpennessParametersReleased
Institut Teknologi Bandung (ITB)FT-whisper (IDSV)Institut Teknologi Bandung (ITB), Universitas Indonesia, JAIST, NAISTspeech recognitionunknown–11 Oct 2024Institut Teknologi Bandung (ITB)MMS fine-tuned, no LM (IDSV)Institut Teknologi Bandung (ITB), Universitas Indonesia, JAIST, NAISTspeech recognitionunknown–11 Oct 2024Institut Teknologi Bandung (ITB)MMS fine-tuned with LM (IDSV)Institut Teknologi Bandung (ITB), Universitas Indonesia, JAIST, NAISTspeech recognitionunknown–11 Oct 2024

Scores

BenchmarkModelScoreWho ran itSetting
IDSV (Indonesian Speech Variability)FT-whisper (IDSV)10.38 WERindependentIDSV group RFC
IDSV (Indonesian Speech Variability)FT-whisper (IDSV)13.04 WERindependentIDSV group SIM
IDSV (Indonesian Speech Variability)FT-whisper (IDSV)13.12 WERindependentIDSV group RFM
IDSV (Indonesian Speech Variability)FT-whisper (IDSV)14.85 WERindependentIDSV test, all groups
IDSV (Indonesian Speech Variability)FT-whisper (IDSV)15.98 WERindependentIDSV group SFM
IDSV (Indonesian Speech Variability)FT-whisper (IDSV)29.64 WERindependentIDSV group SIC
IDSV (Indonesian Speech Variability)FT-whisper (IDSV)5.44 CERindependentIDSV test, all groups
IDSV (Indonesian Speech Variability)MMS fine-tuned with LM (IDSV)19.95 WERindependentIDSV test, all
IDSV (Indonesian Speech Variability)MMS fine-tuned with LM (IDSV)6.45 CERindependentIDSV test, all
IDSV (Indonesian Speech Variability)MMS fine-tuned, no LM (IDSV)29.66 WERindependentIDSV test, all
IDSV (Indonesian Speech Variability)MMS fine-tuned, no LM (IDSV)7.86 CERindependentIDSV test, all
Where this comes from
Every recorded score of this organisation's models, with who ran it. Scores from different settings are not directly comparable.

Sources

  1. Enhancing Indonesian Automatic Speech Recognition: Evaluating Multilingual Models with Diverse Speech Variabilities arxiv.org https://arxiv.org/html/2410.08828v2

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "JAIST", https://aurion.semai.id/organisations/jaist/, records of October 2026.