Automatic Speech Recognition (Latest)
Automatic Speech Recognition (Latest)

Customization

This section provides some examples on how to customize Riva ASR to boost specific words at runtime with word boosting.

Word boosting allows you to bias the ASR engine to recognize particular words of interest at request time by giving them a higher score when decoding the output of the acoustic model.

Copy
Copied!
            

python3 python-clients/scripts/asr/transcribe_file.py --server 0.0.0.0:50051 --input-file en-US_sample.wav --language-code en-US --boosted-lm-words "nature" --boosted-lm-score 30

Above command will print the transcript as shown below.

Copy
Copied!
            

## what is nature language processing

The service is able to generate the transcript with nature instead of natural in this example.

Previous Configuration
Next Performance
© Copyright © 2024, NVIDIA Corporation. Last updated on Aug 6, 2024.