AssemblyAI introduced significant improvements to API's inference latency, processing most audio files under 45 seconds.
New Real-Time-Factor (RTF) as low as .008x enables faster processing of lengthy audio files without compromising accuracy.
Cost reductions initiated for both asynchronous and real-time speech-to-text models, reflecting efficiency gains.
Intelligent mini batching and hardware parallelization techniques boost processing throughput on GPUs.
Increased default concurrency limits for async and real-time services, along with reduced pricing for several Audio Intelligence models.

AssemblyAI is an AI software company that offers state-of-the-art speech recognition and natural language processing services. They specialize in transcribing audio and video files with high accuracy, providing solutions that cater to developers and businesses looking to integrate speech-to-text capabilities into their services. Their API enables easy integration, handling various accents and noise conditions effectively. The company predominantly focuses on making spoken information more accessible and actionable, offering insights through sentiment analysis, content moderation, and topic detection, thus helping organizations to enhance customer experiences, automate processes, and derive valuable data from spoken audio.

AssemblyAI

Advanced Speech AI for analyzing and summarizing voice data efficiently.

AssemblyAI Enhances API with Faster Speeds and Reduced Costs

Recent Product News