Model Overview

This collection contains large size versions of Conformer-CTC (around 120M parameters) trained on ULCA & Europal with around ~2900 hours. The model transcribes speech in hindi characters along with spaces for hinglish speech.

Model Architecture

Conformer-CTC model is a non-autoregressive variant of Conformer model [1] 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

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.

The checkpoint of the language model used as the neural rescorer can be found here. You may find more info on how to train and use language models for ASR models here: ASR Language Modeling

Datasets

All the models in this collection are trained on Hindi labelled dataset(~2900 hrs):

a. ULCA Hindi Corpus

b. Europal Dataset

Tokenizer Construction

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 \
 --log

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 and 6-Gram KenLM trained on AI4Bharat Corpus and Europal.

*- Normalized and without special characters and punctuation.

**- KenLM with 128 beam size with n_gram_alpha=1.0, n_gram_beta=1.0.

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 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.EncDecCTCModelBPE.from_pretrained(model_name="stt_hi_conformer_ctc_large")
output_transcript = asr_model.transcribe([filenames])

Transcribing text with this model

python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py \
    pretrained_name="stt_hi_conformer_ctc_large \
    audio_dir="" \
    dataset_manifest="" \
    output_filename="" \
    batch_size=32 \
    cuda=0 \

Transcribing text using buffered/chunked streaming with this model

python [NEMO_GIT_FOLDER]/examples/asr/asr_chunked_inference/ctc/speech_to_text_buffered_infer_ctc.py \
 --asr_model="stt_hi_conformer_ctc_large" \
 --test_manifest="" \
 --output_path="" \
 --model_stride=4

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.

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 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.