Assemblyai Reduces Streaming Speech-to-Text Pricing by 68% from $0.47/hr to $0.15/hr, Enhancing Accessibility for Users

Claude 3.7 Sonnet LLM model added to LeMUR offerings

BAA for HIPAA Compliance moved from Custom plan only to Pay-as-you-go and Custom plans (feature democratization)

Streaming Speech-to-Text feature 'End of Utterance Detection' renamed to 'End of Turn Detection'

Keyterm Prompting feature added to Slam-1 model (supports up to 1000 domain-specific words or phrases)

Streaming Speech-to-Text decreased from $0.47/hr to $0.15/hr (68% reduction)

Free tier pre-recorded audio hours decreased from 416 hours to 185 hours

Streaming concurrency limit changed from '100 streams' to '5 new streams per minute' for Free tier, with Pay-as-you-go now starting at '100 concurrent streams with 10% ramp per minute'

Pre-recorded Speech-to-Text restructured from Best/Nano tiers to Slam-1/Universal models, both now priced at $0.27/hr (Best decreased from $0.37, Nano increased from $0.12)

Free tier streaming audio hours limit added: Up to 333 hours

Claude 3.5 Haiku LLM model added to LeMUR offerings