Exploiting its third-party eXtension Partner Program, Bluetooth proponent CSR has enabled its BlueCore5-Multimedia platform to support integrated speech synthesis, also known as text-to-speech (TTS), and speech recognition functionality. Developed in co-operation with NowSpeak Technologies and Rubidium Ltd., CSR now offers headset and hands-free car kit users a true hands-free solution using the BlueCore5-Multimedia platform. According to CSR, BlueCore5 silicon offers an integrated 64 MIPs DSP and on-chip memory to host third-party software that enables new features and functions when using Bluetooth.
By combining speech synthesis and speech recognition technologies, CSR provides users with control when communicating via either a Bluetooth headset or embedded car kit, said the developer. When a call is received on the user’s mobile phone, the speech synthesis function announces the name or number of the caller, allowing the user to decide whether or not to accept the call. Speech recognition is used to answer the call without the need to press a button. Speech synthesis can be used as a connection and pairing status guide and can inform the user of the title of a music track played over Bluetooth. Furthermore, speech recognition allows users to easily and safely control a vast array of functions through voice commands, including call acceptance and rejection, voice dialing, and music control and selection, according to CSR.
In this solution, NowSpeak offers the complete hands-free interface for CSR BlueCore devices with its next-generation text-to-speech and speech recognition software. Its text-to-speech technology automatically synthesizes speech from text in real time and has been specifically optimized to speak people's names, phone numbers and device status phrases with a natural voice. With support for all major languages and a small memory footprint it is possible to include multiple languages in a Bluetooth device. Its speaker independent speech recognition requires no user training, supports multiple vocabularies and can be used to recognize words, digits or whole phrases. NowSpeak's speech recognition is designed to work in real world situations with background noise reduction and rejection of unwanted sounds. According to NowSpeak, its text-to-speech and speech recognition solutions can be integrated together or used independently.
Likewise, Rubidium’s speech synthesis and speech recognition is also available on the BlueCore5-Multimedia platform. Rubidium’s speech synthesis uses concatenate TTS based on real human speech to produce pure and natural sound, while keeping use of computational resources exceptionally low. It supports diverse languages, and was designed to offer rapid and flexible addition of virtually any language and voice. Rubidium’s core speech recognition technology is derived from its proprietary Rubidium Dialog Engine series and features speaker independent recognition and supports keyword spotting and noise robustness. Optimized to support essentially any language and accent, it uses artificial intelligence elements to cope with natural language, dynamic vocabulary and multi-layer dialogues.