Scribe enables complete customization of its language model and acoustic model using custom data.
Customizing the Language Model
The language model that Scribe uses will have an impact on the words and phrases recognized from an audio sample. If your audio contains words or phrases that Scribe's language model does not know (such as proper names), these will not transcribe accurately.
GreenKey Scribe's ScribeTrain container allows you to customize the transcription language model from any word set. You can use this feature to add completely new words or phrases or provide representative text to increase the accuracy of target phrases. Language model customization requires launching both the transcription container and the ScribeTrain container.
We recommend a machine with at least 8 GB of RAM and 4 CPUs for building custom language models. Our ScribeTrain documentation discusses how to perform language model customization.
Customizing the Acoustic Model
If Scribe poorly transcribes a particular voice sample or accent, then the acoustic model is likely poorly trained on that type of voice. An example of this is two speakers speaking the same phrase, but possessing different accents and resulting in different accuracy transcriptions.
Below are a couple of ways to improve the acoustic model.
GreenKey has a voice onboarding process that you can participate in. Visit this page, then simply press start, and read the 6 phrases shown (hitting next between each phrase). At the end a ".wav" and ".stm" file will be downloaded by the page. Email these two files to firstname.lastname@example.org. The files are completely anonymized.
Once received, we will inform you in about a week how you can use an updated model with improved accuracy for the voice samples you send.
Custom Acoustic Model Training
GreenKey Scribe does allow custom acoustic model training for firms that have GPU computing capabilities. Check out the documentation for ScribeTrain.