101AITools
    Submit a tool
    Back to directory
    Open-source text-to-speech / voice cloning
    Freemium
    Free Trial
    Sponsored

    Coqui

    Open-source text-to-speech / voice cloning

    Coqui was an open-source voice AI company best known for 🐸TTS, a deep-learning toolkit for text-to-speech, and XTTS-v2, a multilingual voice-cloning model. The company announced shutdown in late 2023 and went offline in early 2024, but the codebase, models, and community fork remain widely used for local TTS, voice cloning, and research. XTTS-v2 supports 17 languages and can clone a voice from ~6 seconds of reference audio with streaming inference under ~200 ms in optimized setups.

    Related guides

    Browse curated shortlists where this tool appears.

    All use cases · All industries

    Alternatives & similar tools

    Related picks from editorial notes.

    Also mentioned

    • ElevenLabs — Cloud TTS/voice cloning with commercial licensing and API.
    • PlayHT — SaaS voice generation and cloning for production apps.
    • Piper / Kokoro — Fully permissive open-source TTS for commercial local deployment.

    Features & details

    Full overview from our catalog (read-only reference).

    Category (short)
    Open-source text-to-speech / voice cloning
    Category
    Open-source text-to-speech / voice cloning
    Pricing (CSV)
    Free Trial
    Directory pricing
    Freemium
    Sponsored note
    no
    Target audience
    Developers building local/offline TTS pipelines, Researchers experimenting with voice cloning and multilingual speech, Indie creators prototyping voice features on GPU/CPU hardware, Not ideal for: teams needing vendor support, SLAs, or straightforward commercial licensing for XTTS-v2 weights.
    Best for
    Developers building local/offline TTS pipelines, Researchers experimenting with voice cloning and multilingual speech, Indie creators prototyping voice features on GPU/CPU hardware, Not ideal for: teams needing vendor support, SLAs, or straightforward commercial licensing for XTTS-v2 weights.
    Pricing notes
    Verified (July 2026): Library code — MPL 2.0, free and open source (GitHub: github.com/coqui-ai/TTS). XTTS-v2 model weights — Coqui Public Model License (CPML); non-commercial use only. Former commercial license — Coqui previously sold XTTS commercial licenses (~$365/yr cited by third parties); no longer available after shutdown. Hosted Coqui Studio/API — Discontinued when company shut down (Jan 2024). Community package — `pip install coqui-tts` (community-maintained fork).

    Pros & cons

    Editorial notes to help compare fit before opening the vendor site.

    Pros

    • High-quality multilingual voice cloning from minimal reference audio
    • Fully local/offline deployment possible
    • Large ecosystem, docs, and Hugging Face integration
    • Active community after company shutdown
    • Multiple model architectures in one toolkit

    Cons

    • Coqui Inc. shut down — no vendor support or commercial license sales
    • XTTS-v2 weights restricted to non-commercial use under CPML
    • Setup requires Python/GPU expertise vs plug-and-play SaaS
    • Company-hosted Studio/API no longer available
    • Long-term roadmap depends on community, not a funded vendor

    Review notes

    Context from the listing review and editorial research.

    Coqui/XTTS is widely regarded as one of the best open-source voice-cloning stacks, with strong community adoption (XTTS hit ~1M downloads quickly per founder posts). The company closure is the main caveat: no commercial support, CPML blocks commercial XTTS output without a separate license path, and future model development depends on community effort. Developers praise quality and flexibility; legal/licensing complexity is the top enterprise concern.

    Extended features

    In-depth description and capability notes.

    XTTS-v2 multilingual voice cloning — 17 languages; cross-language voice transfer from short audio clips. Open-source TTS library — Tacotron2, VITS, YourTTS, XTTS backends via Python (`coqui-tts` / GitHub). Streaming inference — Low-latency speech generation for assistants and real-time apps. Voice conversion — Zero-shot voice conversion alongside TTS in the same framework. Hugging Face model hub — Pre-trained weights (e.g. coqui/XTTS-v2) for local deployment. Community maintenance — Original repo coqui-ai/TTS; active community fork after company closure.

    Use cases

    Coqui fits developers, researchers, and hobbyists who need self-hosted TTS or voice cloning without cloud API costs. Common workflows: audiobook narration, video localization, accessibility tools, conversational AI prototypes, and academic speech synthesis. Production commercial use of XTTS-v2 weights requires careful license review (CPML restricts commercial use; the company that sold commercial licenses no longer exists).