No official torrent yet — requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from nvidia/parakeet-tdt_ctc-110m
All models by nvidia →
|
|
parakeet-tdt_ctc-110m is an ASR model that transcribes speech with Punctuations and Capitalizations of the English alphabet.
This model is jointly developed by NVIDIA NeMo and Suno.ai teams.
It is a Large version of Hybrid FastConformer [1] TDT-CTC [2] (around 114M parameters) model.
This model has been trained with fastconformer architecture with full attention hence this model can transcribe up to 20 minutes of audio in one single pass.
This model achieves super fast RTFx values across all datasets, with an average RTFx of ~5300 on Hugging Face leaderboard evaluation sets on A100.
See the model architecture section and NeMo documentation for complete architecture details.
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 latest PyTorch version.
The model is available for use in the NeMo toolkit [3], and can be used as a pre-trained checkpoint for inference or for fine-tuning on another dataset.
First, let's get a sample
Then simply do:
By default model uses TDT to transcribe the audio files, to switch decoder to use CTC, use decoding_type='ctc'
This model accepts 16000 Hz mono-channel audio (wav files) as input.
This model provides transcribed speech as a string for a given audio sample.
This model uses a Hybrid FastConformer-TDT-CTC architecture. FastConformer [1] is an optimized version of the Conformer model with 8x depthwise-separable convolutional downsampling. You may find more information on the details of FastConformer here: Fast-Conformer Model.
The NeMo toolkit [3] was used for finetuning this model for 20,000 steps over parakeet-tdt-1.1 model. This model is trained with this example script and this base config.
The tokenizers for these models were built using the text transcripts of the train set with this script.
The model was trained on 36K hours of English speech collected and prepared by NVIDIA NeMo and Suno teams.
The training dataset consists of private subset with 27K hours of English speech plus 9k hours from the following public PnC datasets:
The performance of Automatic Speech Recognition models is measured using Word Error Rate. Since this dataset is trained on multiple domains and a much larger corpus, it will generally perform better at transcribing audio in general.
The following table summarizes the performance of the available models in this collection with the Transducer decoder. Performances of the ASR models are reported in terms of Word Error Rate (WER%) with greedy decoding.
These are greedy WER numbers without external LM. More details on evaluation can be found at HuggingFace ASR Leaderboard
NVIDIA Riva, is an accelerated speech AI SDK deployable on-prem, in all clouds, multi-cloud, hybrid, on edge, and embedded. Additionally, Riva provides:
Although this model isn’t supported yet by Riva, the list of supported models is here.
Check out Riva live demo.
[1] Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition
[2] Efficient Sequence Transduction by Jointly Predicting Tokens and Durations
[3] Google Sentencepiece Tokenizer
[5] Suno.ai
[6] HuggingFace ASR Leaderboard
[7] Towards Measuring Fairness in AI: the Casual Conversations Dataset
License to use this model is covered by the CC-BY-4.0. By downloading the public and release version of the model, you accept the terms and conditions of the CC-BY-4.0 license.
pip install nemo_toolkit['all']import nemo.collections.asr as nemo_asrasr_model = nemo_asr.models.ASRModel.from_pretrained(model_name="nvidia/parakeet-tdt_ctc-110m")wget https://dldata-public.s3.us-east-2.amazonaws.com/2086-149220-0033.wavoutput = asr_model.transcribe(['2086-149220-0033.wav'])print(output[0].text)python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py pretrained_name="nvidia/parakeet-tdt_ctc-110m" audio_dir="<DIRECTORY CONTAINING AUDIO FILES>"