NVIDIA NeMo is a conversational AI toolkit built for researchers working on automatic speech recognition (ASR), natural language processing (NLP), and text-to-speech synthesis (TTS). The primary objective of NeMo is to help researchers from industry and academia to reuse prior work (code and pretrained models and make it easier to create new conversational AI models.

Introductory video.

• Speech processing

  • Automatic Speech Recognition (ASR)

    • Supported models: Jasper, QuartzNet, CitriNet, Conformer-CTC, Conformer-Transducer, ContextNet, ...
    • Supports CTC and Transducer/RNNT losses/decoders
    • Beam Search decoding
    • Language Modelling for ASR: N-gram LM in fusion with Beam Search decoding, Neural Rescoring with Transformer

  • Speech Classification and Speech Command Recognition: MatchboxNet (Command Recognition)
  • Voice activity Detection (VAD): MarbleNet
  • Speaker Recognition: SpeakerNet, ECAPA_TDNN
  • Speaker Diarization: SpeakerNet
  • Pretrained models on different languages.: English, Spanish, German, Russian, Chinese, French, Italian, Polish, ...
  • NGC collection of pre-trained speech processing models.

• Natural Language Processing

  • Compatible with Hugging Face Transformers and NVIDIA Megatron
  • Neural Machine Translation (NMT)
  • Punctuation and Capitalization
  • Token classification (named entity recognition)
  • Text classification
  • Joint Intent and Slot Classification
  • BERT pre-training
  • Question answering
  • GLUE benchmark
  • Information retrieval
  • Entity Linking
  • Dialogue State Tracking
  • Neural Duplex Text Normalization
  • NGC collection of pre-trained NLP models.

• Speech synthesis (TTS)

  • Spectrogram generation: Tacotron2, GlowTTS, FastSpeech2, FastPitch, FastSpeech2
  • Vocoders: WaveGlow, SqueezeWave, UniGlow, MelGAN, HiFiGAN
  • End-to-end speech generation: FastPitch_HifiGan_E2E, FastSpeech2_HifiGan_E2E
  • NGC collection of pre-trained TTS models.

• Tools

  • Text Processing (text normalization and inverse text normalization)
  • CTC-Segmentation tool
  • Speech Data Explorer: a dash-based tool for interactive exploration of ASR/TTS datasets

Built for speed, NeMo can utilize NVIDIA's Tensor Cores and scale out training to multiple GPUs and multiple nodes.

1. Python 3.6, 3.7 or 3.8
2. Pytorch 1.8.1 or above
3. NVIDIA GPU for training

Version	Status	Description
Latest		Documentation of the latest (i.e. main) branch.
Stable		Documentation of the stable (i.e. most recent release) branch.

A great way to start with NeMo is by checking one of our tutorials.

FAQ can be found on NeMo's Discussions board. You are welcome to ask questions or start discussions there.

Use this installation mode if you want the latest released version.

apt-get update && apt-get install -y libsndfile1 ffmpeg
pip install Cython
pip install nemo_toolkit['all']

Use this installation mode if you want the a version from particular GitHub branch (e.g main).

apt-get update && apt-get install -y libsndfile1 ffmpeg
pip install Cython
python -m pip install git+https://github.com/NVIDIA/NeMo.git@{BRANCH}#egg=nemo_toolkit[all]

Use this installation mode if you are contributing to NeMo.

apt-get update && apt-get install -y libsndfile1 ffmpeg
git clone https://github.com/NVIDIA/NeMo
cd NeMo
./reinstall.sh

Note that RNNT requires numba to be installed from conda.

conda remove numba
pip uninstall numba
conda install -c numba numba

If you chose to work with main branch, we recommend using NVIDIA's PyTorch container version 21.08-py3 and then installing from GitHub.

docker run --gpus all -it --rm -v <nemo_github_folder>:/NeMo --shm-size=8g \
-p 8888:8888 -p 6006:6006 --ulimit memlock=-1 --ulimit \
stack=67108864 --device=/dev/snd nvcr.io/nvidia/pytorch:21.08-py3

Many example can be found under "Examples" folder.

We welcome community contributions! Please refer to the CONTRIBUTING.md CONTRIBUTING.md for the process.

NeMo is under Apache 2.0 license.