Skip to content
HUIAHUB

Topic

Speech, TTS & STT

Speech recognition, text-to-speech and voice cloning: transcription tools, open models, diarization and the guides for running them.

Speech AI covers speech-to-text (STT), which transcribes spoken audio, and text-to-speech (TTS), which turns written text into a synthetic voice, along with voice cloning. To start with speech AI, read the guide to transcribing audio locally with Whisper, or the guide to open-source text to speech you can run yourself with Piper, Kokoro and other engines.

The tool pages cover hosted and desktop products: ElevenLabs for text to speech, voice cloning and dubbing, Descript for editing recordings through their transcript, and Mac apps for dictation and meeting memory. Muse Gadgets, Meta's open-source SDKs and firmware for devices that connect to its Muse assistant, is listed too. VoiceStudio, an open-source desktop app, runs voice cloning, speech generation, dubbing and transcription on your own computer.

The videos cover Gemini text to speech with voice cloning and speaker diarization with Nemotron 3. Choose a local guide or VoiceStudio when audio must stay on your machine, and compare pricing labels on the tool cards when a hosted service fits.