This collection contains large size versions of Conformer-CTC (around 120M parameters) trained on NeMo ASRSet with over 1500 hours of French speech. Includes two models wtih separate tokenization sets, along with accompanying language models (n=4).
Utilizes a Google SentencePiece  tokenizer with vocabulary size 128, and transcribes text in lower case French alphabet along with spaces, apostrophes, and hyphens. Secondary model and associated language model (indicated with "no_hyphen" infix) omits hyphen from tokenization
Conformer-CTC model is a non-autoregressive variant of Conformer model  for Automatic Speech Recognition which uses CTC loss/decoding instead of Transducer. You may find more info on the detail of this model here: Conformer-CTC Model.
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 over a thousand hours of French speech:
Both models use same dataset, excluding a preprocessing step to strip hyphen from data for secondary model's training.
(Note: The MultiliingualLibreSpeech set includes orthography of both pre- and post- 1990 reform, which may affect predictability of hyphenation.)
The tokenizer for this model was built using text corpus provided with the train dataset.
We build a token set with the following script:
python [NEMO_GIT_FOLDER]/scripts/tokenizers/process_asr_text_tokenizer.py \ --manifest="train_manifest.json" \ --data_root="" \ --vocab_size=128 \ --tokenizer="spe" \ --spe_type="unigram" \ --spe_character_coverage=1.0 \ --no_lower_case \ --log
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
With 128 beam search and 4gram KenLM model (included with this model):
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. For the secondary model trained without hyphenation, scores are slightly improved (see version sheet).
The model is available for use in the NeMo toolkit , 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.EncDecCTCModelBPE.from_pretrained(model_name="stt_fr_conformer_ctc_large")
python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py \ pretrained_name="stt_fr_conformer_ctc_large" \ audio_dir=""
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.
Further, since portions of the training set contain text from both pre- and post- 1990 orthographic reform, regularity of punctuation may vary between the two styles. For downstream tasks requiring more consistency, finetuning or downstream processing may be required. If exact orthography is not necessary, then using secondary model is advised.