This collection contains the extra extra large(XXL) size version of Fast Conformer-Transducer (around 1.1B parameters) trained on NeMo ASRSet with around 24500 hours of english speech. The model transcribes speech in lower case English alphabet along with spaces and apostrophes.
Fast Conformer-Transducer model is an autoregressive variant of Conformer model [1] for Automatic Speech Recognition which uses Transducer loss/decoding. You may find more info on the detail of this model here: Fast-Conformer Model.
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.
All the models in this collection are trained on a composite dataset (NeMo ASRSET) comprising of several thousand hours of English speech:
1.20.1 version model is first pretrained on Librilight 60k hrs of data and then finetuned on NeMo ASR Set 3.0
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 | MCV Test 8.0 | Train Dataset |
---|---|---|---|---|---|---|---|---|---|---|
1.20.0 | SentencePiece Unigram | 1024 | 3.04 | 1.59 | 1.27 | 2.13 | 5.84 | 4.88 | 6.45 | NeMo ASRSET 3.0 |
1.20.1 | SentencePiece Unigram | 1024 | 2.71 | 1.50 | 1.09 | 2.00 | 4.48 | 4.32 | 5.11 | NeMo ASRSET 3.0 |
You may use language models to improve the accuracy of the models.
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.
import nemo.collections.asr as nemo_asr
asr_model = nemo_asr.models.EncDecRNNTModelBPE.from_pretrained(model_name="stt_en_fastconformer_transducer_xxlarge")
python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py \
pretrained_name="stt_en_fastconformer_transducer_xxlarge" \
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.
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.
[1] Conformer: Convolution-augmented Transformer for Speech Recognition
[2] Google Sentencepiece Tokenizer
License to use this model is covered by the NGC TERMS OF USE 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 NGC TERMS OF USE.