Alternatives to Naomi

Voicebox is an open-source desktop AI voice studio for local voice cloning, dictation, and agent speech — no account required, MIT licensed.

OpenClaw Voice is a free, MIT-licensed, self-hosted voice chat interface for talking to your AI assistants from any browser.

OmniVoice Studio is an open-source desktop studio for local voice cloning, voice design, dubbing, dictation and audiobook production.

Independent guide and web console for ByteDance Seed voice models, run through a server-side KIE audio API.

Krisp pairs real-time AI noise cancellation with a bot-free AI notetaker, accent conversion, and voice translation for calls.

AI audio platform: text to speech, voice cloning, transcription and phone/chat voice agents on one shared credit pool.

Voiceitt is inclusive speech recognition that makes non-standard speech work with dictation, captions and voice assistants

MiniMax Audio text-to-speech API: speech-2.8 synthesis with 10-second voice cloning and per-character pricing.

Smallest.ai delivers realtime text-to-speech, speech-to-text, speech-to-speech and voice agents on one developer API.

Fish Audio delivers expressive AI text-to-speech and 15-second voice cloning across 30+ languages with emotion tags.

Supertonic: free on-device text-to-speech that runs locally through ONNX Runtime, with no per-character cost.

On-device speech SDK for real-time ASR, voice cloning, and full-duplex voice agents that never touch the cloud

Apple's free tool that turns your finished ePub into an audiobook using its own digital narration voices.

WeConnect is a multilingual social discovery app for meeting friends across borders with real-time chat translation.

Browser-based PDF, EPUB, DOCX and TXT to MP3 or chaptered M4B audiobook converter with a text-review step.

Check how strong your English accent sounds to a US listener in about 30 seconds — free, in your browser, no signup.

Open-source Vietnamese text-to-speech voice cloning you download and self-host on your own GPU

ComfyUI VoxCPM runs open-source multilingual text-to-speech with voice cloning and voice design on your own GPU.

An MIT-licensed, self-hosted AI runtime that serves text, voice, vision, image, video, 3D and agent workloads behind an OpenAI-compatible API.

Free open-source Japanese text-to-speech models you self-host, six MioTTS sizes from 0.1B to 2.6B.

Free open-source multilingual speech synthesizer that gives screen reader users a native-language voice, running fully offline on Windows, Android, and Linux.

Open-source instant voice cloning from a 10-second clip, with separate control knobs for emotion, accent, rhythm, pauses, and intonation.

Open-source port of the Xiaozhi AI voice chatbot to embedded Linux single-board computers

Free, open-source TTS web app and OpenAI-compatible API using Microsoft Edge's neural voices.

QVAC SDK: one open-source API for on-device AI across text, vision and audio, with no cloud

Open-source Japanese text-to-speech with emoji-driven style control you can fine-tune

Open-source text-to-speech with fine-grained emotion and prosody control

Cboard is a free open-source AAC web app that speaks symbol boards and typed text in 46 languages.

Hume AI measures 600+ dimensions of vocal and facial expression to evaluate whether voice AI actually feels understood.

Realtime TTS, streaming STT, speech-to-speech and open-model inference on one credit balance and one API key.