nvidia/stt_it_fastconformer_hybrid_large_pc

NVIDIA FastConformer-Hybrid Large (it)

| | |

This model transcribes speech in upper and lower case Italian alphabet along with spaces, periods, commas, and question marks. It is a "large" version of FastConformer Transducer-CTC (around 115M parameters) model. This is a hybrid model trained on two losses: Transducer (default) and CTC. See the model architecture section and NeMo documentation for complete architecture details.

NVIDIA NeMo: Training

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.

pip install nemo_toolkit['all']

How to Use this Model

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.

Automatically instantiate the model

import nemo.collections.asr as nemo_asr
asr_model = nemo_asr.models.EncDecHybridRNNTCTCBPEModel.from_pretrained(model_name="nvidia/stt_it_fastconformer_hybrid_large_pc")

Transcribing using Python

First, let's get a sample

wget https://dldata-public.s3.us-east-2.amazonaws.com/2086-149220-0033.wav

Then simply do:

asr_model.transcribe(['2086-149220-0033.wav'])

Transcribing many audio files

Using Transducer mode inference:

python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py
 pretrained_name="nvidia/stt_it_fastconformer_hybrid_large_pc"
 audio_dir="<DIRECTORY CONTAINING AUDIO FILES>"

Using CTC mode inference:

python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py
 pretrained_name="nvidia/stt_it_fastconformer_hybrid_large_pc"
 audio_dir="<DIRECTORY CONTAINING AUDIO FILES>"
 decoder_type="ctc"

Input

This model accepts 16000 Hz Mono-channel Audio (wav files) as input.

Output

This model provides transcribed speech as a string for a given audio sample.

Model Architecture

FastConformer [1] is an optimized version of the Conformer model with 8x depthwise-separable convolutional downsampling. The model is trained in a multitask setup with joint Transducer and CTC decoder loss. You may find more information on the details of FastConformer here: Fast-Conformer Model and about Hybrid Transducer-CTC training here: Hybrid Transducer-CTC.

Training

The NeMo toolkit [3] was used for training the models for over several hundred epochs. These model are 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.

Datasets

The model in this collection are trained on a composite dataset (NeMo PnC IT ASRSET) comprising of 487 hours of Italian speech:

• Mozilla Common Voice 12.0 (Italian) - 220 hours after data cleaning. Speech Data Processor config used to prepare this data is here.
• Multilingual LibriSpeech (Italian) - 214 hours after data cleaning. Speech Data Processor config used to prepare this data is here.
• VoxPopuli transcribed subset (Italian) - 53 hours after data cleaning. Speech Data Processor config used to prepare this data is here.

Performance

The performance of Automatic Speech Recognition models is measuring 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 tables 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.

a) On data without Punctuation and Capitalization

Version	Tokenizer	Vocabulary Size	MCV 12.0 Dev	MCV 12.0 Test	MLS Dev	MLS Test	VoxPopuli Dev	VoxPopuli Test
1.20.0	SentencePiece BPE	512	5.19%	5.64%	13.01%	11.39%	13.02%	16.22%

b) On data with Punctuation and Capitalization

Version	Tokenizer	Vocabulary Size	MCV 12.0 Dev	MCV 12.0 Test	MLS Dev*	MLS Test*	VoxPopuli Dev	VoxPopuli Test
1.20.0	SentencePiece BPE	512	7.70%	8.11%	21.69%	18.27%	16.96%	19.97%

* We use only a subset of dev/test sets with P&C restored from the original books

Limitations

Since this model was trained on publically available speech datasets, the performance of this model might degrade for speech which includes technical terms, or vernacular that the model has not been trained on. The model might also perform worse for accented speech. The model only outputs the punctuations: '.', ',', '?' and hence might not do well in scenarios where other punctuations are also expected.

NVIDIA Riva: Deployment

NVIDIA Riva, is an accelerated speech AI SDK deployable on-prem, in all clouds, multi-cloud, hybrid, on 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

Although this model isn’t supported yet by Riva, the list of supported models is here. Check out Riva live demo.

References

[1] Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition

[2] Google Sentencepiece Tokenizer

[3] NVIDIA NeMo Toolkit

Licence

License to use this model is covered by the CC-BY-4 License unless another License/Terms Of Use/EULA is clearly specified. By downloading the public and release version of the model, you accept the terms and conditions of the CC-BY-4 License.