This model transcribes speech in lowercase English alphabet including spaces and apostrophes, and is trained on several thousand hours of English speech data. It is a non-autoregressive "large" variant of Streaming Citrinet, with around 140 million parameters. See the model architecture section and NeMo documentation for complete architecture details. It is also compatible with NVIDIA Riva for production-grade server deployments.
The model is available for use in the NeMo toolkit , and can be used as a pre-trained checkpoint for inference or for fine-tuning on another dataset.
To train, fine-tune or play with the model you will need to install NVIDIA NeMo. We recommend you install it after you've installed the latest PyTorch version.
pip install nemo_toolkit['all']
import nemo.collections.asr as nemo_asr asr_model = nemo_asr.models.EncDecCTCModelBPE.from_pretrained("nvidia/stt_en_citrinet_1024_gamma_0_25")
First, let's get a sample
Then simply do:
python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py pretrained_name="nvidia/stt_en_citrinet_1024_gamma_0_25" audio_dir="<DIRECTORY CONTAINING AUDIO FILES>"
This model accepts 16000 kHz Mono-channel Audio (wav files) as input.
This model provides transcribed speech as a string for a given audio sample.
Streaming Citrinet-1024 model is a non-autoregressive, streaming variant of Citrinet model  for Automatic Speech Recognition which uses CTC loss/decoding instead of Transducer. You may find more info on this model here: Citrinet Model.
The tokenizer for this models was built using the text transcripts of the train set with this script.
All the models in this collection are trained on a composite dataset (NeMo ASRSET) comprising of several thousand hours of English speech:
- Librispeech 960 hours of English speech
- Fisher Corpus
- Switchboard-1 Dataset
- WSJ-0 and WSJ-1
- National Speech Corpus (Part 1, Part 6)
Note: older versions of the model may have trained on smaller set of datasets.
The list of the available models in this collection is shown in the following table. Performances of the ASR models are reported in terms of Word Error Rate (WER%) with greedy decoding.
|Version||Tokenizer||Vocabulary Size||LS test-other||LS test-clean||WSJ Eval92||WSJ Dev93||NSC Part 1||Train Dataset|
|1.0.0||SentencePiece Unigram||1024||7.6||3.4||2.5||4.0||6.2||NeMo ASRSET 1.0|
While deploying with NVIDIA Riva, you can combine this model with external language models to further improve WER. The WER(%) of the latest model with different language modeling techniques are reported in the following table.
Since this model was trained on publicly available speech datasets, the performance of this model might degrade for speech that includes technical terms, or vernacular that the model has not been trained on. The model might also perform worse for accented speech.
For the best real-time accuracy, latency, and throughput, deploy the model with NVIDIA Riva, an accelerated speech AI SDK deployable on-prem, in all clouds, multi-cloud, hybrid, at the edge, and embedded. Additionally, Riva provides:
- World-class out-of-the-box accuracy for the most common languages with model checkpoints trained on proprietary data with hundreds of thousands of GPU-compute hours
- Best in class accuracy with run-time word boosting (e.g., brand and product names) and customization of acoustic model, language model, and inverse text normalization
- Streaming speech recognition, Kubernetes compatible scaling, and Enterprise-grade support Check out Riva live demo.
- Downloads last month