First question
What happens to the recording?
Separate transcription, voice generation, realtime agents, and meeting notes before comparing the tools that touch spoken data.
Theme
AI Resources
Voice, speech, and meeting AI
A focused map for turning speech into text, text into voices, and live conversations into AI workflows.
Voice tools can involve sensitive recordings or speaker likeness. Check consent, storage, source terms, and data handling before using real conversations.
Voice boundary
Separate speech from speaker likeness
A tool that transcribes audio is not the same as one that generates, clones, or performs a voice. Check the source terms for the actual use.
Testing path
Start with low-risk audio
Use sample clips or non-sensitive meetings first, then check retention, export, deletion, and review controls before using important recordings.
Last updated
July 9, 2026
Use this page to narrow what to explore next, not as an endorsement, recommendation, guarantee, or safety review. Open the original source before relying on setup, terms, limits, privacy, access, or costs.
Transcribe and understand audio
Speech-to-text and audio understanding projects
6Use this group when the job is turning speech, noisy audio, calls, or recordings into text or structured context.
Cohere Transcribe
CohereLabs/cohere-transcribe-03-2026
A 2B parameter automatic speech recognition model for audio-in, text-out transcription across 14 languages.
Mega-ASR
xzf-thu/Mega-ASR
A robust automatic speech recognition project for messy real-world audio, with code, inference and training paths, Hugging Face weights, a technical report, Voices-in-the-Wild-2M, and project-reported results across difficult acoustic scenarios.
MiMo-V2.5-ASR
XiaomiMiMo/MiMo-V2.5-ASR
A Xiaomi MiMo speech-recognition model focused on Mandarin, English, Chinese dialects, code-switched speech, noisy audio, songs, and multi-speaker transcription.
MOSS-Audio
OpenMOSS/MOSS-Audio
An audio-understanding model family for speech, sound, music, captioning, time-aware QA, ASR, and reasoning over real-world audio.
NVIDIA Nemotron 3.5 ASR Streaming 0.6B
nvidia/nemotron-3.5-asr-streaming-0.6b
An NVIDIA 600M-parameter multilingual streaming automatic speech recognition model for low-latency voice AI and high-throughput transcription, with a Hugging Face model card, NeMo usage paths, language-locales, performance tables, and OpenMDW license terms to inspect.
insanely-fast-whisper
Vaibhavs10/insanely-fast-whisper
An opinionated CLI for very fast on-device transcription with Whisper.
Generate or run voices
Text-to-speech and voice generation projects
9Open this group when the output is a spoken voice, a compact TTS model, or a voice workflow that needs source and consent checks.
Fish Audio S2 Pro
fishaudio/s2-pro
A text-to-speech model with detailed control over prosody and emotional delivery.
KittenTTS
KittenML/KittenTTS
A very small text-to-speech model designed to stay lightweight without feeling toy-like.
Kokoro-82M
hexgrad/Kokoro-82M
A compact 82M-parameter text-to-speech model from hexgrad, with model facts, usage examples, voice materials, samples, a demo Space, and a linked GitHub inference library.
MOSS-TTS Family
OpenMOSS/MOSS-TTS
A speech and sound generation model family covering TTS, voice design, spoken dialogue, realtime speech, compact speech generation, and MOSS-SoundEffect-v2.0 text-to-audio materials.
MOSS-TTS-Nano
OpenMOSS/MOSS-TTS-Nano
A tiny multilingual speech generation model positioned for real-time TTS, CPU-friendly local use, and lightweight deployment.
sarashina2.2-tts
sbintuitions/sarashina2.2-tts
A Japanese-centric text-to-speech system from SB Intuitions, with Japanese and English generation, style transfer, and zero-shot voice generation support.
Supertonic
supertone-inc/supertonic
An on-device multilingual text-to-speech system built around ONNX Runtime, with local inference, 31-language support, expression tags, model assets, and examples across browser, mobile, desktop, and edge runtimes.
VoxCPM2
openbmb/VoxCPM2
A multilingual text-to-speech model with voice design, controllable voice cloning, and streaming support.
Voicebox
jamiepine/voicebox
A local-first voice synthesis studio for voice cloning, speech generation, effects, and voice-powered app workflows.
Meetings and realtime agents
Live conversation and meeting workflows
6Use this group when timing, rooms, calls, summaries, or meeting records are the hard part.
LiveKit Agents
livekit/agents
A realtime framework for voice, video, and physical AI agents, with Python and Node.js paths, LiveKit room participants, WebRTC clients, telephony support, tools, testing, and deployment options.
Pipecat
pipecat-ai/pipecat
A Python framework and ecosystem for real-time voice and multimodal AI agents, with audio/video pipelines, transports, client SDKs, structured flows, and subagent support.
June
open-software-network/os-june
An open source desktop AI assistant for meeting notes, dictation, projects, and local agent sessions, with a Tauri app, June API backend, local app state by default, macOS and Windows release paths, and TEE verification notes for the backend.
Meetily
Zackriya-Solutions/meetily
A local-first AI meeting assistant with public source, desktop release assets for Windows and macOS, Linux source-build docs, live transcription, meeting summaries, local Whisper or Parakeet transcription, local or external LLM summary options, optional analytics, and an MIT license.
OpenOats
yazinsai/OpenOats
A meeting note-taking assistant designed to be more conversational and responsive than passive transcription.
PersonaPlex
NVIDIA/personaplex
A real-time full-duplex speech-to-speech conversational model with persona control through role prompts and voice conditioning.
Also in AI
Follow the next layer.
Keep the thread going with AI Guides for decision habits for messy AI choices, AI Access for free and low-cost ways to compare AI model access, AI Ballot for a clearer view of what readers are leaning toward.