On August 24, 2026, an analysis characterized Pika Labs as excelling in speed and personality for social-first clips through features like Pikaffects and Pikaframes, though it noted limitations for creating longer content. Earlier, on August 18, Pika Labs introduced several new audio capabilities, including Pika Speech, which generates voices from five seconds of reference audio for up to five minutes per request. The company also launched Pika SFX, a model that creates production-ready sound effects in real-time from written descriptions, supporting both single events and complex sequences with specified timing, environment, texture, and mood. Additionally, it unveiled Pika Soundtrack, which generates audio that understands event occurrences and timing. These releases followed the August 14 introduction of Pika Audio, presented as a suite of four frontier audio models.
Subscribe for full access to Pika Labs's products in full detail 
