The FIRST Company to Develop Multimodal AI with 200+ Installations Worldwide
The FIRST Company to Develop Multimodal AI with 200+ Installations Worldwide
AimeSpeech : Speech recognition (STT), speech synthesis (TTS), Speaker identification
AimeSpeech is a core speech processing framework inside the Aimenicorn software ecosystem. AimeSpeech includes a Speech Recognition engine (Speech to Text, STT engine), a Speech Synthesis engine (Text to Speech, TTS engine), Speaker identification library and other advanced speech processing libraries. AimeSpeech supports English, Japanese, Vietnamese, Korean. It can be run both on-premise and on-cloud. It works with PC/Mac and also smartphones.
AimeSpeech is applied in various Multimodal AI products of Aimesoft, such as AimeHotel, AimeReception, Aime AIShop, ...
Please use Google Chrome browser on PC/Mac (not on smartphones) to view demo Web pages that use AimeSpeech below. For smartphones, you need to use AimeSpeech SDK/API for integration.
AimeSpeech Features
-
Speech recognition (Speech-to-Text, STT) -
Emotion analysis -
Speech Synthesis (Text-to-Speech, TTS) -
Speaker identification -
Sentiment Analysis from Speech features -
Supports English, Japanese, Vietnamese
AimeSpeech Technologies
Aimesoft’s technologies to realize AimeSpeech
-
Speech Recognition TechnologiesDNN (Deep Neural Networks), LSTM, GMM, HMM, Acoustic modelling, Language model -
Natural Language Processing technologiesTokenization, POS tagging, keyword extraction, word normalization -
Speech Synthesis TechnologiesHMM, Festival, Kaldi, Tacotron2, DNN, Vocoder -
Speaker Identification TechnologiesGMM, VGG network
AimeSpeech Benefits
Various applications of AimeSpeech
-
Call center automation
-
Multimodal identification, authentication (using both face and voice)
-
AimeBank : bank user authentication
bản demo TTS