You searched a paid product

Open-source alternatives to ElevenLabs

ElevenLabs does 4 jobs: speech generation, voice cloning, speech recognition, voice agents. Most projects replace one or two of them, so pick the job you actually need.

Quick answer

Facts checked 16 Oct 2026

Open-source alternatives to ElevenLabs that run locally, allow commercial use and need no paid API include GPT-SoVITS (speech generation and voice cloning; MIT; runs on CPU), Dia (speech generation and voice cloning; Apache-2.0; GPU, ~4.4GB+) and whisper.cpp (speech recognition; MIT; runs on CPU). None of the 22 projects we track covers all 4 ElevenLabs jobs, so most people combine two or more.

22 projects

Bolna

bolna-ai/bolna

★ 795 active today

Open-source orchestration framework for LLM voice agents that make and take phone calls over Twilio or Plivo.

Replaces Voice agents (partly) Not: speech generation, voice cloning, speech recognition

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · MIT
Paid services: No: Required
Hardware: Unknown: Unknown
Setup: Confirmed: Medium

Compare 6 of 8 facts sourced · checked today GitHub ↗ View details

Chatterbox

resemble-ai/chatterbox #3 TRENDING

★ 27k active 2mo ago

Resemble AI's family of open TTS models (Turbo, Nano, Multilingual) with zero-shot voice cloning from a reference clip.

Replaces Speech generation + Voice cloning (partly) Not: speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · MIT
Paid services: Partial: Optional only
Hardware: Partial: CPU works
Setup: Confirmed: Easy

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Coqui TTS

coqui-ai/TTS

★ 46k active 2y ago

Python TTS library with many models (XTTS-v2, VITS, YourTTS, Bark, Tortoise); original repo unmaintained, continued by idiap fork.

Replaces Speech generation + Voice cloning (partly) Not: speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Partial: Conditions · MPL-2.0
Paid services: Confirmed: None required
Hardware: Unknown: Unknown
Setup: Confirmed: Easy

Compare 7 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Dia

nari-labs/dia

★ 19k active 11mo ago

1.6B-parameter English text-to-dialogue model from Nari Labs with two-speaker tags, non-verbal sounds and audio-prompt voice cloning.

Replaces Speech generation + Voice cloning (partly) Not: speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · Apache-2.0
Paid services: Confirmed: None required
Hardware: Confirmed: GPU, ~4.4GB+
Setup: Confirmed: Easy

Compare 7 of 8 facts sourced · checked today GitHub ↗ View details

F5-TTS

SWivid/F5-TTS

★ 15k active 2d ago

Flow-matching text-to-speech research model that clones a voice zero-shot from a reference audio clip.

Replaces Speech generation + Voice cloning (partly) Not: speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: No: No · MIT code, CC-BY-NC-4.0 weights
Paid services: Confirmed: None required
Hardware: Unknown: Unknown
Setup: Confirmed: Medium

Compare 7 of 8 facts sourced · checked 1d ago GitHub ↗ View details

faster-whisper

SYSTRAN/faster-whisper

★ 26k active today

Reimplementation of OpenAI Whisper on CTranslate2, up to 4x faster with less memory, with int8 on CPU or GPU.

Replaces Speech recognition Not: speech generation, voice cloning, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · MIT
Paid services: Confirmed: None required
Hardware: Confirmed: CPU works
Setup: Confirmed: Easy

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Fish Speech

fishaudio/fish-speech #4 TRENDING

★ 33k active 2d ago

Fish Audio's multilingual TTS and voice cloning stack (Drama 3, S2-Pro) under a source-available community licence.

Replaces Speech generation + Voice cloning Not: speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Partial: Conditions · Custom licence
Paid services: Partial: Optional only
Hardware: Confirmed: GPU, ~24GB+
Setup: Confirmed: Medium

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

GPT-SoVITS

RVC-Boss/GPT-SoVITS

★ 63k active 8d ago

Few-shot voice cloning and TTS WebUI; 5-second zero-shot or 1-minute fine-tune, five languages, MIT code and weights.

Replaces Speech generation + Voice cloning (partly) Not: speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · MIT
Paid services: Confirmed: None required
Hardware: Partial: CPU works
Setup: Confirmed: Easy

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Kokoro

hexgrad/kokoro #5 TRENDING

★ 9.4k active 14mo ago

Lightweight 82M-parameter open-weight text-to-speech model with a pip-installable Python inference library.

Replaces Speech generation (partly) Not: voice cloning, speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · Apache-2.0
Paid services: Confirmed: None required
Hardware: Unknown: Unknown
Setup: Confirmed: Easy

Compare 7 of 8 facts sourced · checked 1d ago GitHub ↗ View details

LiveKit Agents

livekit/agents

★ 15k active today

Python framework for real-time voice agents as server-side participants in LiveKit WebRTC rooms, with STT/LLM/TTS plugins.

Replaces Voice agents (partly) Not: speech generation, voice cloning, speech recognition

Runs locally: Confirmed: Yes
Commercial use: Partial: Conditions · Apache-2.0
Paid services: No: Required
Hardware: Partial: CPU works
Setup: Confirmed: Easy

Compare 8 of 8 facts sourced · checked today GitHub ↗ View details

MeloTTS

myshell-ai/MeloTTS

★ 7.7k active 22mo ago

Multilingual text-to-speech library from MyShell.ai and MIT covering English accents, Spanish, French, Chinese, Japanese and Korean.

Replaces Speech generation (partly) Not: voice cloning, speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · MIT
Paid services: Confirmed: None required
Hardware: Partial: CPU works
Setup: Confirmed: Medium

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

OpenVoice

myshell-ai/OpenVoice

★ 38k active 18mo ago

MyShell's instant voice cloning model with tone-colour transfer, style control and cross-lingual cloning; MIT code and weights.

Replaces Speech generation + Voice cloning (partly) Not: speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · MIT
Paid services: Confirmed: None required
Hardware: Unknown: Unknown
Setup: Confirmed: Medium

Compare 7 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Orpheus TTS

canopyai/Orpheus-TTS

★ 6.4k active 10mo ago

Llama-based 3B speech LLM from Canopy Labs for expressive, emotive, low-latency streaming text-to-speech.

Replaces Speech generation + Voice cloning (partly) Not: speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Partial: Conditions · Apache-2.0
Paid services: Partial: Optional only
Hardware: Partial: CPU works
Setup: Confirmed: Medium

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Pipecat

pipecat-ai/pipecat

★ 17k active today

Open-source Python framework for real-time voice and multimodal agents that plugs together STT, LLM, TTS and transports.

Replaces Voice agents (partly) Not: speech generation, voice cloning, speech recognition

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · BSD-2-Clause
Paid services: Partial: Optional only
Hardware: Unknown: Unknown
Setup: Confirmed: Easy

Compare 6 of 8 facts sourced · checked today GitHub ↗ View details

Piper

OHF-Voice/piper1-gpl

★ 5.9k active 1d ago

Fast, local neural text-to-speech engine with downloadable voices in 40+ languages, used by Home Assistant and NVDA.

Replaces Speech generation (partly) Not: voice cloning, speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Partial: Conditions · GPL-3.0-or-later
Paid services: Confirmed: None required
Hardware: Partial: CPU works
Setup: Confirmed: Easy

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Scriberr

rishikanthc/Scriberr

★ 3.1k active 3w ago

Self-hosted web app for offline transcription with Whisper, Parakeet or Canary, speaker diarization and LLM summaries.

Replaces Speech recognition Not: speech generation, voice cloning, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · MIT
Paid services: Partial: Optional only
Hardware: Partial: CPU works
Setup: Confirmed: Medium

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Sesame CSM

SesameAILabs/csm

★ 15k active 16mo ago

Sesame's 1B Conversational Speech Model that generates speech from text plus optional conversational audio context.

Replaces Speech generation (partly) Not: voice cloning, speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Partial: Conditions · Apache-2.0
Paid services: Confirmed: None required
Hardware: Partial: See details
Setup: Confirmed: Medium

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

StyleTTS 2

yl4579/StyleTTS2

★ 6.4k active 2y ago

Research text-to-speech model using style diffusion and adversarial training, with pretrained LJSpeech and LibriTTS checkpoints.

Replaces Speech generation + Voice cloning (partly) Not: speech recognition, voice agents

Runs locally: Confirmed: Yes
Commercial use: Partial: Conditions · MIT
Paid services: Confirmed: None required
Hardware: Partial: CPU works
Setup: Confirmed: Medium

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

VoiceStudio

debpalash/VoiceStudio #1 TRENDING

★ 70k active today

Local desktop app for voice cloning, voice design, video dubbing, dictation, transcription and audiobooks, built on OmniVoice and other engines.

Replaces Speech generation + Voice cloning + Speech recognition (partly) Not: voice agents

Runs locally: Confirmed: Yes
Commercial use: No: No · AGPL-3.0-only code, CC-BY-NC-4.0 weights
Paid services: Partial: Optional only
Hardware: Confirmed: CPU works
Setup: Confirmed: Easy

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

whisper.cpp

ggml-org/whisper.cpp

★ 54k active 4d ago

Dependency-free C/C++ port of OpenAI's Whisper speech recognition, running on CPU or GPU across desktop, mobile and the browser.

Replaces Speech recognition Not: speech generation, voice cloning, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · MIT
Paid services: Confirmed: None required
Hardware: Confirmed: CPU works
Setup: Confirmed: Hard

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Whisper

openai/whisper #2 TRENDING

★ 111k active 6w ago

OpenAI's general-purpose speech recognition model for multilingual transcription, translation to English and language ID.

Replaces Speech recognition Not: speech generation, voice cloning, voice agents

Runs locally: Confirmed: Yes
Commercial use: Confirmed: Yes · MIT
Paid services: Confirmed: None required
Hardware: Confirmed: GPU, ~1GB+
Setup: Confirmed: Easy

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

WhisperX

m-bain/whisperX

★ 25k active 2w ago

Fast batched Whisper transcription with word-level timestamps via wav2vec2 alignment and pyannote speaker diarization.

Replaces Speech recognition Not: speech generation, voice cloning, voice agents

Runs locally: Confirmed: Yes
Commercial use: Partial: Conditions · BSD-2-Clause
Paid services: Confirmed: None required
Hardware: Confirmed: CPU works
Setup: Confirmed: Medium

Compare 8 of 8 facts sourced · checked 1d ago GitHub ↗ View details

Common questions about ElevenLabs alternatives

Is there a free, open-source alternative to ElevenLabs?

Yes. VoiceStudio, GPT-SoVITS, Fish Speech, Chatterbox and F5-TTS are free to download and run on your own computer without paid API keys. Each covers only part of what ElevenLabs does, so check which job you need.

Can I do voice cloning locally without an ElevenLabs subscription?

Yes, with Fish Speech (Custom licence), GPT-SoVITS (MIT) and Chatterbox (MIT). Check the hardware notes and licence conditions on each page before you start.

Can I do speech recognition locally without an ElevenLabs subscription?

Yes, with whisper.cpp (MIT), faster-whisper (MIT) and WhisperX (BSD-2-Clause). Check the hardware notes and licence conditions on each page before you start.

Which ElevenLabs alternatives can I use commercially?

Confirmed: Bolna, Chatterbox, Dia, faster-whisper, GPT-SoVITS, Kokoro, MeloTTS, OpenVoice, Pipecat, Scriberr, whisper.cpp and Whisper. With conditions: Coqui TTS, Fish Speech, LiveKit Agents, Orpheus TTS, Piper, Sesame CSM, StyleTTS 2 and WhisperX. Not for commercial use: F5-TTS and VoiceStudio, usually because the model weights are non-commercial. This is not legal advice; read each licence.

Does any open-source project replace all of ElevenLabs?

Not as of 2026-10-16. Speech generation, voice cloning, speech recognition and voice agents are covered by different projects.

Open-source alternatives to ElevenLabs

Quick answer

Your requirements

22 projects

Bolna

Chatterbox

Coqui TTS

Dia

F5-TTS

faster-whisper

Fish Speech

GPT-SoVITS

Kokoro

LiveKit Agents

MeloTTS

OpenVoice

Orpheus TTS

Pipecat

Piper

Scriberr

Sesame CSM

StyleTTS 2

VoiceStudio

whisper.cpp

Whisper

WhisperX

Common questions about ElevenLabs alternatives

Cookie preferences