Collecting Sentences for Common Voice

" summary

Introduces Mozilla’s Common Voice project and why large numbers of diverse sentences are needed for an open speech dataset. Describes multiple concrete methods used to collect sentences that volunteers can read for recording (including technical details of the collection mechanisms).

AI-written summary · about

" related talks

• 44mCommon Voice - Building Multilingual Voice DatasetsKalyan Dikshit · foss-north 2019
• 35mProduct fairness testing for Machine Learning developersCherish Molezion, Qazi Mamunur Rashid · Google I/O 2022
• 16mThe importance of representative dataMike Knapp, Emila Yang, Hannah Highfill, Ben Hutchinson · Google I/O 2022
• 31mWhen humans teach machinesShai Hertz · WeAreDevelopers AI Congress Vienna 2018
• 02mis everywhereCypher 2017
• 27mStardust: Digital Tooling for Earthly Science & EducationLiting Chen, Francis P. Crawley · IPFS Camp 2024
• 35mAnalyzing Public Conversation using LDA and Topic Modeling, in PythonSveta Gimpelson · stackconf 2023
• 38mQuestions and DiscussionWrong at the Root: Racial Bias and the Tension Between Numbers and Words in Non-Internet Data