2024 was a busy year at Picovoice with new products, languages, SDKs, pioneering on-device AI research and development to continue to offer best-in-class on-device AI. See the 2024 highlights from Picovoice:
Introducing the only end-to-end Local LLM Platform: picoLLM
In May 2024, Picovoice announced the world's only local LLM platform: picoLLM.
picoLLM is the end-to-end local large language model (LLM) platform that enables AI assistants to run on-device, on-premises, and in the private cloud without sacrificing accuracy.
picoLLM platform consists of picoLLM Inference, the cross-platform Local LLM Inference Engine that supports all LLM architectures, picoLLM Compression, the compression algorithm that reduces runtime and storage requirements of any LLM while retaining model performance, and picoLLM GYM, compression-aware small language model training tool.
Orca Streaming Text-to-Speech
In May 2024, Picovoice added streaming text input processing capability to Orca Text-to-Speech, allowing Orca to enable LLM-powered voice AI agents with very low latency.
Curious to learn how Orca Streaming Text-to-Speech achieves such low latency compared to other voice generators? Learn about Dual Streaming Synthesis.
Eagle Speaker Recognition
In January 2024, Picovoice announced the version 1 release of its Speaker Recognition engine, Eagle.
Developers can improve productivity, save costs, and enhance user experience by identifying speakers using their unique voice characteristics with Eagle Speaker Recognition. Eagle, powered by deep learning, enables developers to determine "who is speaking" on any platform with no language or passphrase limitations.
Multilingual On-device Streaming Speech-to-Text with Cheetah
In December, Picovoice announced that Cheetah Streaming Speech-to-Text supports five new languages: French, German, Italian, Portuguese, and Spanish. This has enabled more developers to build private and cost-effective AI applications with zero network latency using accurate, production-ready, cross-platform Cheetah Streaming Speech-to-Text.
Looking for new languages? Talk to Picovoice Consulting and get a custom model trained for your application when you become an Enterprise Plan customer.
Most Popular Tutorials
- On-device Text-to-Speech in Python
- Real-time Speaker Identification in Python
- How to Run a Local LLM on Android
Most Popular Articles
- Top Free and Open-source Large Language Models
- Top Free and Commercial Speaker Diarization APIs and SDKs
- What is GPTQ?
The Most Popular SDK
Once again, Python was the developers' first choice this year. If you haven't yet already, start building with a few lines of Python:
