Hugging Face's logo

nvidia
/
stt_en_fastconformer_hybrid_large_streaming_multi

Automatic Speech Recognition
NeMo
PyTorch
English
speech
streaming
audio
Transducer
Conformer
CTC
NeMo
Eval Results
Model card Files Community
2
stt_en_fastconformer_hybrid_large_streaming_multi / README.md
vnoroozi's picture
vnoroozi
Update README.md
4fb986e
preview code
|
raw
history blame
13.7 kB
metadata
language:
  - en
library_name: nemo
datasets:
  - librispeech_asr
  - fisher_corpus
  - Switchboard-1
  - WSJ-0
  - WSJ-1
  - National-Singapore-Corpus-Part-1
  - National-Singapore-Corpus-Part-6
  - vctk
  - VoxPopuli-(EN)
  - Europarl-ASR-(EN)
  - Multilingual-LibriSpeech-(2000-hours)
  - mozilla-foundation/common_voice_8_0
  - MLCommons/peoples_speech
thumbnail: null
tags:
  - automatic-speech-recognition
  - speech
  - streaming
  - audio
  - Transducer
  - Conformer
  - CTC
  - pytorch
  - NeMo
license: cc-by-4.0
widget:
  - example_title: Librispeech sample 1
    src: https://cdn-media.huggingface.co/speech_samples/sample1.flac
  - example_title: Librispeech sample 2
    src: https://cdn-media.huggingface.co/speech_samples/sample2.flac
model-index:
  - name: stt_en_fastconformer_hybrid_large_streaming_multi
    results:
      - task:
          type: Automatic Speech Recognition
          name: automatic-speech-recognition
        dataset:
          name: LibriSpeech (other)
          type: librispeech_asr
          config: other
          split: test
          args:
            language: en
        metrics:
          - name: Test WER
            type: wer
            value: 5.4

NVIDIA Streaming Conformer-Hybrid Large (en-US)

| Model architecture | Model size | Language

This collection contains large size versions of cache-aware FastConformer-Hybrid (around 114M parameters) with multiple look-ahead support, trained on a large scale english speech. These models are trained for streaming ASR which be used for streaming applications with a variety of latencies. All models are hybrid with both Transducer and CTC decoders. See the model architecture section and NeMo documentation for complete architecture details.

Model Architecture

FastConformer [4] is an optimized version of the Conformer model [1]. 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. You may find more on how to switch between the Transducer and CTC decoders in the documentations.

These models are cache-aware versions of Hybrid FastConfomer which are trianed for streaming ASR. You may find more info on cache-aware models here: Cache-aware Streaming Conformer. The models are trained with multiple look-aheads which makes the model to be able to support different latencies. To learn on how to switch between different look-aheads, you may read the documentations on the cache-aware models.

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 SentencePiece tokenizers [2] for these models were built using the text transcripts of the train set with this script.

Datasets

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)
  • VCTK
  • VoxPopuli (EN)
  • Europarl-ASR (EN)
  • Multilingual Librispeech (MLS EN) - 2,000 hours subset
  • Mozilla Common Voice (v7.0)
  • People's Speech - 12,000 hrs subset

Note: older versions of the model may have trained on smaller set of datasets.

Performance

The list of the available models in this collection is shown in the following tables for both CTC and Transducer decoders. Performances of the ASR models are reported in terms of Word Error Rate (WER%) with greedy decoding.

Transducer Decoder

Version Tokenizer Vocabulary Size att_context_sizes LS test-other ([70,13]-1040ms) LS test-other ([70,6]-480ms) LS test-other ([70,1]-80ms) LS test-other ([70,0]-0s) Train Dataset
1.20.0 SPE Unigram 1024 [[70,13],[70,6],[70,1],[70,0]] 5.4 5.7 6.4 7.0 NeMo ASRSET 3.0

CTC Decoder

Version Tokenizer Vocabulary Size att_context_sizes LS test-other ([70,13]-1040ms) LS test-other ([70,6]-480ms) LS test-other ([70,1]-80ms) LS test-other ([70,0]-0s) Train Dataset
1.20.0 SPE Unigram 1024 [[70,13],[70,6],[70,1],[70,0]] 6.2 6.7 7.8 8.4 NeMo ASRSET 3.0

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 streaming 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 latest Pytorch version.

pip install nemo_toolkit['all']
'''
'''
(if it causes an error): 
pip install nemo_toolkit[all]

You may use this script to simulate streaming ASR with these models: cache-aware streaming simulation. You may use --att_context_size to set the context size otherwise the default which is the first context size in the list is going to be used.

Automatically load the model from NGC 

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

Transcribing text with this model

Using Transducer mode inference:

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

Using CTC mode inference:

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

To change between different look-aheads you may set att_context_size of the script transcribe_speech.py as the following:

python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py \
  pretrained_name="stt_en_fastconformer_hybrid_large_streaming_multi" \
  audio_dir="<DIRECTORY CONTAINING AUDIO FILES>" \
  att_context_size=[70,0]

Input

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

Output

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

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.

References

[1] Conformer: Convolution-augmented Transformer for Speech Recognition

[2] Google Sentencepiece Tokenizer

[3] NVIDIA NeMo Toolkit

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

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']
'''
'''
(if it causes an error): 
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.EncDecRNNTBPEModel.from_pretrained("nvidia/stt_en_conformer_transducer_xlarge")

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 

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

Input

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

Output

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

Model Architecture

Conformer-Transducer model is an autoregressive variant of Conformer model [1] for Automatic Speech Recognition which uses Transducer loss/decoding instead of CTC Loss. You may find more info on the detail of this model here: Conformer-Transducer Model.

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

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)
  • VCTK
  • VoxPopuli (EN)
  • Europarl-ASR (EN)
  • Multilingual Librispeech (MLS EN) - 2,000 hrs subset
  • Mozilla Common Voice (v8.0)
  • People's Speech - 12,000 hrs subset

Note: older versions of the model may have trained on smaller set of datasets.

Performance

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 MLS Test MLS Dev MCV Test 8.0 Train Dataset
1.10.0 SentencePiece Unigram 1024 3.01 1.62 1.17 2.05 5.70 5.32 4.59 6.46 NeMo ASRSET 3.0

Limitations

Since this model was trained on publicly 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.

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] Conformer: Convolution-augmented Transformer for Speech Recognition [2] Google Sentencepiece Tokenizer [3] NVIDIA NeMo Toolkit

Licence

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.