Custom tuning

Table 1. Custom tuning features for Voci's ASR
FeatureReal-time/Post CallLocal/V‑CloudLanguagesDescription
Word substitutions BothBothAllVoci's AutoSubs process to automatically identify words that are typically transcribed incorrectly.
Auto-substitutionsBothBothAllA proprietary process where Voci analyzes a sample of a customer's audio and associated transcripts in order to systematically identify words/phrases that are typically transcribed incorrectly, and as a result will generate a substitutions list for your use case to improve transcription accuracy.
Language modelsBothBothAllThe ASR engine uses machine learning components known as models to represent knowledge about speech. This knowledge is applied during transcription. The two types of models used are acoustic models and language models. The acoustic model converts audio into a stream of sound symbols specific to a language, such as English or Spanish. The language model is responsible for converting the stream of sound symbols into text.
Custom language modelsBothBothAllCreating a custom language model improves overall accuracy and adds new words to the model dictionary that are specific to a customer's business.
Custom acoustic modelsBothBothAllIn "Full" custom language and acoustic modeling, the same custom language modeling process is supplemented with the creation of a custom acoustic model that is specifically attuned to the customer's call acoustics.
Custom speed tuningBothBothAllVoci provides services to custom tailor latency and throughput needs based on required SLA's and turnaround times needed.