Google's new 4B Gemma model cuts speaker-labelling errors on all four benchmarks, two barely

T2Covered by 1 source · first reported by UploadVR / Road to VR / MIXED at 05 Oct, 11:05 UTC

A 4B-parameter model published under Google's Hugging Face organisation on October 4 reports lower speaker-labelling error on all four of the standard diarisation test sets, with the improvements marked as statistically significant at p < 0.0001. Google is blunt about what it has released: "This is…

iotreleasestelecomxr

Sources

• T2Google's new 4B Gemma model cuts speaker-labelling errors on all four benchmarks, two barelyUploadVR / Road to VR / MIXED05 Oct, 11:05