Cartesia launches Multilingual Voices to preserve speaker identity across languages
Summary
Cartesia has introduced its new Multilingual Voices feature, which allows companies to maintain a consistent speaker identity across different languages while preserving aspects such as tone and pacing. This innovation addresses a significant challenge in speech translation, where the identity of the speaker can often be lost due to variations in rhythm and emphasis across languages. Cartesia's Sonic 3.6 model, which currently ranks first on both the Provider Voice and Controlled Voice TTS leaderboards, utilizes state space models for efficient and scalable speech inference, ensuring high-quality voice performance in multilingual applications.
Analysis
Nvidia: Nvidia is a technology company focused on GPUs and AI infrastructure. It participated as an investor in Cartesia's funding round alongside other venture firms. Cartesia: Cartesia develops speech synthesis technology using state space models for efficient, low-latency AI voice generation. The company recently launched Multilingual Voices, a feature that maintains consistent speaker identity, tone, and character across languages for business applications. It partners with local businesses like Baklavastory to demonstrate voice consistency in customer interactions. Krandiash: Krandiash is affiliated with Cartesia and introduced the Multilingual Voices feature in a public announcement. The announcement highlighted partnerships and the technical challenges of preserving voice identity across languages. Lightspeed: Lightspeed is a venture capital firm that backs early-stage technology companies. It participated in Cartesia's funding round alongside Nvidia and other investors. Baklavastory: Baklavastory is a business based in the Mission district that collaborated with Cartesia to showcase Multilingual Voices. The partnership demonstrates how the technology maintains brand voice consistency for customer-facing applications across languages. Index Ventures: Index Ventures is a venture capital firm specializing in technology and innovation investments. It joined Cartesia's funding round to back the company's AI voice technology advancements. Kleiner Perkins: Kleiner Perkins is a venture capital firm that invests in technology companies. It participated in Cartesia's recent funding round supporting the development of advanced speech models. Model Performance: Cartesia's Sonic 3.6 model leads Artificial Analysis leaderboards in voice quality evaluations based on blind comparisons. AI Voice Consistency: Maintaining a single speaker identity across languages addresses a key challenge in speech translation where rhythms and emphasis often shift. Infrastructure Approach: Cartesia relies on state space models to enable scalable, low-latency speech inference suitable for enterprise use.
Categories
aitech