Unverified70% confidenceFactExact time
Google's Universal Speech Model (USM) supports more than 100 languages.
2
Sources
70%
Confidence
Long-term
Relevance
5/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedFLEURS is a multilingual speech dataset released by Google, covering 102 languages.76% similarUnverifiedAudioPaLM, released by Google in 2023, was the first to inject speech tokens directly into a language model's vocabulary, enabling joint modeling of text and audio in the same representation space.68% similarUnverifiedGoogle's goal is to build models that can understand languages exactly as they are expressed, going beyond traditional text translation.67% similarUnverifiedGoogle launched the BERT pre-trained language model in 2018 and the PaLM model with 540 billion parameters in 2022.65% similarUnverifiedGoogle possesses the world's largest accumulation of multilingual speech data from YouTube, search voice input, and the Android system, constituting a moat at the training data level.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/10840API
curl https://kongchang.com/api/v1/knowledge/claims/10840MCP
get_claim(id=10840)