Deepgram

AssemblyAI is a voice AI infrastructure platform that provides APIs and models for transcribing, understanding, and interacting with speech. Its products include pre-recorded, realtime, and synchronous speech-to-text APIs; speech understanding for capabilities such as speaker identification, sentiment, chapters, and summaries; inline audio and transcript guardrails; an LLM Gateway with fallback; and a Voice Agent API with turn detection and interruption handling. The platform can combine speech-to-text, an LLM, and text-to-speech into a real-time voice-agent stack, and supports voice applications such as notetakers, call analytics, medical transcription, dictation, agent assist, and AI scribes. AssemblyAI states that its speech-to-text APIs support 99 languages and that its infrastructure processes about 2 million hours of audio per day.

Google DeepMind is an artificial-intelligence research and development organization that builds AI models and systems for general use, science, robotics, speech, image and video generation, music, and coding. Its featured systems include the Gemini family, Gemma open models, Veo video generation, Lyria music generation, Gemini Robotics, and AI systems for scientific discovery. The site also publishes research and provides access to models through products and developer platforms such as Google AI Studio, the Gemini API, and related enterprise tools.

AssemblyAI's Voice Agent API is an API for building real-time voice agents. It connects speech-to-text, an LLM, and text-to-speech through a single WebSocket, with built-in turn detection and interruption handling.

Speech To Speech is an open-source, modular voice-agent pipeline from Hugging Face. It chains voice activity detection, speech-to-text, a language model, and text-to-speech as separate stages connected by queues; each stage has interchangeable backends. The language-model stage supports OpenAI-compatible protocols and can use hosted providers, Hugging Face Inference Providers, vLLM, or llama.cpp, while the system streams transcripts, generated text, tool calls, and synthesized audio. The project exposes the core OpenAI Realtime event set through WebSocket and WebRTC and includes commands for running a realtime server, a microphone-and-speaker client, or a local setup. It is distributed as a Python package and requires Python 3.10 or newer. The repository says the pipeline is used as the conversation backend for Reachy Mini robots.

Voicebox is a local-first, open-source AI voice studio developed in the jamiepine/voicebox repository. It combines voice cloning, text-to-speech, speech-to-text dictation, voice profiles, audio effects, and agent voice output in a desktop application. It generates speech through seven switchable TTS engines, including Qwen3-TTS, Qwen CustomVoice, LuxTTS, Chatterbox, HumeAI TADA, and Kokoro, and supports voice cloning from reference audio, preset voices, multilingual generation, automatic text chunking, crossfading, and multi-track story editing. Whisper-based transcription powers global dictation, in-app recording, and the transcription API; an optional local Qwen3 LLM can refine dictation or rewrite text according to a voice profile's persona. The application exposes REST endpoints for speech generation, agent speech, transcription, and profile management, plus a built-in MCP server with tools for speaking, transcribing, and browsing captures and profiles. It runs models and stores voice data locally, using Tauri with a React/TypeScript frontend and FastAPI backend; platform backends include MLX on Apple Silicon and PyTorch on CUDA, ROCm, XPU, DirectML, or CPU. The repository is licensed under the MIT License and provides macOS, Windows, and Docker distribution, with Linux build-from-source instructions.