Releases·Tool Release
AssemblyAI launches Universal 3.6 Pro, streaming speech-to-text tuned for voice agents
AssemblyAI launches Universal 3.6 Pro streaming speech-to-text tuned for voice agents
1 engagement1 sourceTue, Sep 29, 2026, 09:04 PM
AI Overview
- AssemblyAI released Universal 3.6 Pro, a streaming speech-to-text model pitched as tuned for voice agent use cases
- The model debuted via a Show HN post linking to assemblyai.com on 2026-09-29
Entities
Related
Releases·Model Release
Microsoft Releases VibeVoice-ASR-Streaming Model on Hugging Face
Microsoft releases VibeVoice-ASR-Streaming
Microsoft releases VibeVoice-ASR-Streaming
175 engagement1 sourceredditThu, Sep 3, 2026, 01:40 AM
Releases·Tool Release
STT-MCP: local speech-to-text tool for AI agents
A developer released STT-MCP, a local speech-to-text tool that enables AI agents to process speech from audio and video files. The project is hosted on GitHub and targets agent workflows needing offline transcription.
A developer released STT-MCP, a local speech-to-text tool that enables AI agents to process speech from audio and video files. The project is hosted on GitHub and targets agent workflows needing offline transcription.
2 engagement1 sourcehackernewsThu, Jul 23, 2026, 03:06 PM
Products·launched
Fusion-runtime – self-hosted voice agents running STT, LLM and TTS in one process
Fusion-runtime is an open-source runtime for building self-hosted voice agents that combine speech-to-text, an LLM, and text-to-speech in a single process. The LLM serves as the conversational reasoning layer between transcription and speech synthesis. It targets developers who want voice agents without depending on external cloud APIs or multiple separate services.
Fusion-runtime is an open-source runtime for building self-hosted voice agents that combine speech-to-text, an LLM, and text-to-speech in a single process. The LLM serves as the conversational reasoning layer between transcription and speech synthesis. It targets developers who want voice agents without depending on external cloud APIs or multiple separate services.
2 engagement1 sourcehackernewsMon, Sep 21, 2026, 04:19 PM
Products·launched
StreamCore – open-source realtime voice infrastructure for AI
StreamCore is an open-source server providing realtime voice infrastructure for AI applications. It handles the audio streaming layer so developers building LLM-powered voice agents can connect speech to models without building transport infrastructure themselves. It targets developers building voice AI products.
StreamCore is an open-source server providing realtime voice infrastructure for AI applications. It handles the audio streaming layer so developers building LLM-powered voice agents can connect speech to models without building transport infrastructure themselves. It targets developers building voice AI products.
1 engagement1 sourcehackernewsFri, Aug 28, 2026, 02:38 PM
Releases·Tool Release
Whisper Live - A nearly-live implementation of Open AI's Whisper
A GitHub project called Whisper Live provides a nearly-live implementation of OpenAI's Whisper speech recognition model. The project enables real-time transcription with low latency, making it useful for live captioning and voice interfaces.
A GitHub project called Whisper Live provides a nearly-live implementation of OpenAI's Whisper speech recognition model. The project enables real-time transcription with low latency, making it useful for live captioning and voice interfaces.
0 engagement1 sourceredditSat, Jul 25, 2026, 04:46 PM