By registering a speech recognition dictionary, it is possible to improve the recognition accuracy of technical terms, etc. and to perform correct text conversion.
The voice recognition API provided by the voice recognition development platform "AmiVoice Cloud Platform" of Advanced Media Co., Ltd. (Headquarters: Toshima-ku, Tokyo, Representative Director, Chairman and President: Kiyoyuki Suzuki) MOBI VOICE (Headquarters: Shinagawa-ku, Tokyo, President: Tomohiro Ishii), an AI telephone automatic answering system
https://mobilus.co.jp/solution/voice
)” was added to “Ver_1.23.0” which added new features.
Dictionary registration to the speech recognition engine can now be done for each domain from the "MOBI VOICE" management screen. This allows you to improve conversion accuracy by registering keywords with low recognition accuracy, such as technical terms and product names, or to register keywords that you want to convert to the official pronunciation and have them converted to the correct keyword. Please note that voice recognition dictionary registration is a paid optional feature.
■About voice recognition dictionary registration function
If you subscribe to the "Voice Recognition Dictionary Registration" option, you can now register your voice recognition dictionary by going to "Settings" > "Voice Recognition Dictionary" on the "MOBI VOICE" management screen. If there are keywords with low recognition accuracy, you can expect to improve the conversion accuracy by registering the "word" and "pronunciation".
Also, by registering the keywords that you want to convert to the official reading, you can convert them to the correct keywords. "Speech recognition dictionary" can be set for each question.
[AmiVoice Cloud Platform, overview]
The speech recognition API provided by AmiVoice Cloud Platform is a development tool for implementing speech recognition functions (real-time recognition/batch recognition) on websites, web applications, smartphone apps, etc. It can be used for various voice recognition services such as voice transcription, voice dialogue, and voice control. AmiVoice has No. 1 share in the voice recognition market
■
is a speech recognition engine that is strong for spoken words. It can be used for a wide range of purposes, including transcription of meetings, voice input, and transcription of telephone audio.
https://acp.amivoice.com/main/
[MOBI VOICE, Overview]
MOBI VOICE is a system that uses voice recognition and speech synthesis engines to automatically respond to telephone inquiries 24 hours a day, 365 days a year. After logging in to the management screen, you can launch an automatic telephone answering service in as little as 5 minutes, allowing you to receive orders and procedures, automatically respond to original scenarios and IVR, and make calls and send information automatically. Outbound calls can also be made.
We will help improve customer satisfaction and service quality by streamlining the response to inquiries at call centers, local government consultation centers, and restaurants and stores, where there is a serious labor shortage due to hiring difficulties. Up to 1,000 cases can be handled simultaneously. Even in the event of a flood of calls from customers or residents due to service failures, disasters, or the new coronavirus, automatic voice scenarios can be created instantly according to the situation, so it can also be used for BCP support.
https://mobilus.co.jp/solution/voice
*Source: ITR “ITR Market View: AI Market 2020” Voice recognition market sales share by vendor (2015-2020 forecast)
Above
