Suddenly, it’s a problem.
The correct answer is… “ Readspeaker ” of…!
Then, the following problem.
The correct answer is “ 1, 2, 3, 4 everything “!
Even if we do not know the product name “ Readspeaker “, the “ Readspeaker “, which is heard from home appliances, the narration of the television program, the narration of the TV program, the “ voice I hear on a daily basis.
“ Readspeaker ” is HOYA Corporation (hereafter HOYA), which is developed from the beginningspeech synthesis technology ( TTS = text to speech ). More than 1,700 companies in Japan alone have more than 1,700 companies, and the introduced companies include “ JR East ” “ Dentsu “, “ Sharp ” “ Kyoto University ” “ Tokyo There are many leading companies, governments, and educational institutions, such as “Tokyo “.
You can immediately understand why it is widely supported by listening to the “ voice “. In addition to the overwhelming “ speech sensation ” and “ clear feeling “, if necessary, “ joy “, “ sad “, “ angry “, etc. Completion that can express delicate emotions.
The quality that does not make you aware of whether it is a person’s voice or a speech synthesis is heard from everywhere in your daily life.
“ Readspeaker ” supports 30 or more languages (as of March 2022) and is being developed globally. It is said that audio technology using AI is more advanced overseas, but “ Readspeaker ” can also gain a large share there.
The core of the technology is “ AI voice synthesis “.
AI speech synthesis and
In “ AI Voice Synthesis “, the introduction of the cutting-edge machine learning technology “deep learning” improves overall quality, and “emotional expression” has become quite close to the delicate and natural human vocalization. rice field.
With conventional technology, the expressive power of detailed nuances, such as “Aizuchi”, “ending”, and “phrase”, which was difficult to express, has been greatly improved.
In general, a high-spec server has been required for the use of “deep learning”, but HOYA’s high technical and development capabilities enable operations on popular terminals. Did.
Readspeaker has a variety of voices from multiple “speakers”, from young women to elderly men.
However, the most advanced technology that does not stay there is a solution that allows you to create “original” voices from the voices of “ Readspeaker ” and the voices of real people (voice actors, actors, celebrities, etc.).
Until now, the recording of the audio required several tens of hours and sometimes over 100 hours. In addition, when adding a new content, the recording was required.
Readspeaker “ Original voice ” allows you to synthesize the audio that makes use of the characteristics of the voice with a recorded voice of about 10 hours, “ joy “, “ angry “,” Raku ** “, you can freely have three emotions. Even if it is not a high-end machine, it is also attractive that such AI speech synthesis can be produced and utilized.
Readspeaker is the destination of AI speech synthesis as well as video games, and is used in various games. As an example, if the character calls under the name set by the player, the immersion will increase, and if it is used in combination with AI, conversation can be conducted.
There is no need to secure voice actors, narrators, and studios for recording when preparing audio, which greatly contributes to man-hours and cost reductions.
If the man-hours are reduced, it will be possible to reflect the audio that mentions the results of the day game on that day in sports games on the same day. In addition, cost reductions have the advantage that even in individuals and small-scale development systems, it makes it easier to direct and express by voice.
Readspeaker’s latest introduction case “Quizknock Stadium”
One of the most recent examples of the introduction of READSPEAKER products is the arcade game “ Quizknock Stadium ” released by Konami Amusement Co., Ltd.
This work is developed with the cooperation of the knowledge group “QuizKnock” led by Takuji Izawa, the University of Tokyo, and is a new sense of quiz game that enjoys communication in virtual space and fast-press quizzes. The operation has started from Thursday).
The questions include problems created by QuizKnock members, and it is also a great attraction that Izawa reads each problem. That’s the last problem here.
The correct answer is “2 speech synthesis using AI **”!
Konami Amusement Co., Ltd. adopted the original voice of Readspeaker in the development of “ Quizknock Stadium “. In the implementation, “ SpeechEngine SDK ” was used.
From the beginning of the development, the company has decided to “add a continuous problem by online updates” and “read out the problem by the University of Tokyo Quiz King Takuji Izawa”. TTS (speech synthesis technology), which can greatly reduce man-hours and costs, solved the challenge of how to achieve it on the busy Izawa’s limited recording schedule.
At first, it was a start from deepening knowledge about speech synthesis technology, but it is used on a daily basis in a wide range of fields and places, and the trust of the solution of Readspeaker and the reproduction of original voice ** The height of the sex has become a decisive factor in the introduction.
Immediately after the introduction, there were times when I misunderstood new words and first appearances that are common in current affairs, but they listed such words and registered as a dictionary each time.
With quiz players nationwide playing “ Quizknock Stadium “, the excellent solution of Readspeaker will be more familiar.
Quizknock Stadium official website
READSPEAKER , which is becoming more and more presence in the future
Not only in the same work, recent video games usually update software via the network and reflect the latest data. In addition, the convergence of Corona due to the new colon virus is still visible, and the record in the studio tends to be delayed today.
Based on such circumstances, TTS (speech synthesis technology) is expected to increase in demand in the game industry, and the voice synthesis solution provided by HOYA , Readspeaker . Will be its powerful candidate.
[READSPEAKER] On the game sound synthesis page, you can listen to “Unity”, “Unreal Engine”, etc., and hear sample sounds. Please take a look.
[READSPEAKER] Game sound voice synthesis page