NVIDIA NeMo Speech
by NVIDIA
Open-source PyTorch toolkit for building and deploying speech recognition, text-to-speech and speech LLM models, with pretrained checkpoints
Voiceover (text-to-speech) Code libraryOpen source
1. Price: github.com/NVIDIA-NeMo/Speech
Free open-source code under Apache 2.0.
github.com/NVIDIA-NeMo/SpeechNo quote yet: price is read from the official pages.
2. Commercial use: LICENSE on GitHub
“each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Work”
You can use it in paid apps and client work.
LICENSE on GitHub3. Credit the author: LICENSE on GitHub
“You must give any other recipients of the Work or Derivative Works a copy of this License”
No credit on screen. Keep the license notice inside the code you ship.
LICENSE on GitHub4. Agent access: docs.nvidia.com/nemo/speech/nightly
An agent can install it by itself with pip install nemo-toolkit.
docs.nvidia.com/nemo/speech/nightlyLicense checked Nov 21, 2026 against 1 official page. Tap a verdict to see its source. How we checked Report an error
NVIDIA NeMo Speech popularity
GitHub stars stayed at 18,775 since Nov 22, 2026, so there is no stars chart.
Downloads from pypistats.org, stars from GitHub, checked Nov 22, 2026. How the Gid Score is calculated
License
Apache License 2.0.
“each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Work” LICENSE on GitHub
“You must give any other recipients of the Work or Derivative Works a copy of this License” LICENSE on GitHub
“You may reproduce and distribute copies of the Work or Derivative Works thereof in any medium, with or without modifications, and in Source or Object form” LICENSE on GitHub
- Keep a copy of the Apache 2.0 license and any NOTICE file when you distribute the code.
- Mark modified files with a prominent notice that you changed them.
- Check the license of each pretrained model checkpoint separately; the repository's Apache 2.0 license covers the code.
- Do not use NVIDIA trademarks or product names beyond describing the origin of the work.
AllowedWith conditionsNot allowedNot stated or not checked yet
Is NVIDIA NeMo Speech free for commercial use? The full answer Not legal advice. Check the license page before you ship.
For your agent
License rules tell your agent what it may do with NVIDIA NeMo Speech. Paste them into your prompt or AGENTS.md.
- NVIDIA NeMo Speech (Voiceover (text-to-speech)). Open-source PyTorch toolkit for building and deploying speech recognition, text-to-speech and speech LLM models, with pretrained checkpoints. Price: free; commercial use OK; no credit needed.
Connect: PyPI nemo-toolkit.
Rule: Keep a copy of the Apache 2.0 license and any NOTICE file when you distribute the code.
Rule: Mark modified files with a prominent notice that you changed them.
Rule: Check the license of each pretrained model checkpoint separately; the repository's Apache 2.0 license covers the code.
Rule: Do not use NVIDIA trademarks or product names beyond describing the origin of the work.
License: https://github.com/NVIDIA-NeMo/Speech/blob/HEAD/LICENSE (checked 2026-11-21; details: https://agentgid.com/assets/nemo/)
Agent installs a package or loads it from a CDN.
pip install nemo-toolkitAlternatives
NVIDIA NeMo Speech This page
Local desktop app for voice cloning, voice design, video dubbing, dictation and transcription in 646 languages, with a local API and an MCP server for agents.
MIT-licensed open TTS with zero-shot cloning in 23 languages and a built-in Perth watermark.
Long-form multi-speaker TTS for podcast-style dialogue, up to 90 minutes and 4 speakers, in English and Chinese.
Zero-shot voice-cloning TTS; code MIT, pre-trained weights CC BY-NC.
Open-source on-device toolkit for speech-to-text and text-to-speech, with models in many languages, for voice apps and agents.