Model Overview
This collection contains ContextNet-1024 (around 140M parameters) trained on French NeMo ASRSet with over 1500 hours of French speech.
It utilizes a Google SentencePiece [1] tokenizer with vocabulary size 1024, and transcribes text in lower case French alphabet along with spaces, apostrophes, and hyphenation.
Model Architecture
ContextNet [2] model is an autoregressive, transducer based Automatic Speech Recognition model. You may find more info on the detail of this model here: ContextNet 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 over fifteen hundred hours of cleaned French speech:
- MCV7.0 356 hours
- MLS 1036 hours
- VoxPopuli 182 hours
Tokenizer Construction
The tokenizer for this model was built using text corpus provided with the train dataset.
We build a Google Sentencepiece Tokenizer [1] with the following script :
python [NEMO_GIT_FOLDER]/scripts/tokenizers/process_asr_text_tokenizer.py \
--manifest="train_manifest.json" \
--data_root="" \
--vocab_size=1024 \
--tokenizer="spe" \
--spe_type="unigram" \
--spe_character_coverage=1.0 \
--no_lower_case \
--log
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 latest model obtains the following greedy scores on the following evaluation datasets -
- 8.32 % on MCV7.0 dev
- 9.42 % on MCV7.0 test
- 6.02 % on MLS dev
- 5.01 % on MLS test
Note that these evaluation datasets have been filtered and preprocessed to only contain French alphabet characters and are removed of punctuation outside of hyphenation and apostrophe.
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 load the model from NGC
import nemo.collections.asr as nemo_asr
asr_model = nemo_asr.models.EncDecRNNTBPEModel.from_pretrained(model_name="stt_fr_contextnet_1024")
Transcribing text with this model
python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py \
pretrained_name="stt_fr_contextnet_1024" \
audio_dir=""
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] Google Sentencepiece Tokenizer
License
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.