ReadSpeaker proudly announces that the two text-to-speech (TTS) engines have topped the Voice Information Associates’ Text-to-Speech Accuracy Testing charts for 2018. The long-time speech industry observer has recognised rSpeak Technologies and Neospeech (which uses ReadSpeaker’s TTS engine) as offering the most accurate TTS voices on the market.
In their freshly released, 2018 Text-to-Speech Accuracy Testing report, Voice Information Associates states that 11 commercially available text to speech products were tested for accuracy performance. Using an English language test corpus of over eighteen hundred phrases, each unit was tested, scored and the results tabulated.
The test corpus (1,588 words/phrases) was segmented into categories which included: Number Handling; Homograph Handling; Handling of Words of Foreign Origin; Acronym Handling; Abbreviation Handling; Name Handling; and Address Handling and the test results were presented in these categories. The average score for all units tested was 80% correct. Overall scores ranged from 60.7% to 98.6% correct.
Among the 11 TTS products tested, representing the text to speech vendors’ latest available releases, rSpeak Technologies stole the show with an overall accuracy rate of 98.6% and Neospeech also proved its outstanding performance at 95.6%.
ReadSpeaker TTS engines leverage a unique combination of text normalization, lexical lookup, grapheme to phoneme modelling, prosody modelling, manuscript creation, recording of voice talent and acoustic database creation to ensure that the outstanding, lifelike quality of ReadSpeaker TTS voices enhances the user experience across all markets.
Please contact us if you want to find out how you can use our premium voices for your business.