Translated from 日本語 • View original In our previous article, we covered Azure Cognitive Services in general. This time, I would like to introduce the voice field service, the content of the service, and the billing system that you are interested in.
What is the Azure Cognitive Services Speech Discipline?
As previously introduced, in the voice field, functions such as "speaking and listening" can be mainly incorporated into the service. You can convert the prepared text into speech and read it aloud, or transcribe the user's voice into a text file. For example, by using the audio field, you can transcribe what the presenter said in the presentation into text data, or conversely, have the prepared manuscript read aloud to the computer. Click here for the official website of the voice field
Services in the Field of Speech
There are four services in the field of voice.
Speech to Text
Speech to Text converts speech to text. This is a function called transcription and dictation. When you enter audio data into Speech to Text, it outputs text data. There are also options that can address needs that cannot be sent to the cloud for voice data. It can also be tuned for use cases where environmental noise is loud, such as construction sites and factories. For example, it can be used to automatically generate a log of what you said in a meeting.
Click here for the official website of Speech to Text
Billing System
Speech to Text's billing structure is pay-as-you-go for what you use. You are charged every hour for the time you convert the audio data you convert to text.
Text to Speech
Text to Speech converts text to speech. This is the so-called computer reading function. When you enter text data into Text to Speech, it outputs audio data. The voice is preset on the service side, but you can also adjust the reading speed and intonation independently to make it smoother and more natural.
For example, it can be used to answer voice chat services, provide audio guidance at reception, and narrate videos.
Click here for the official website of Text to Speech
Billing System
Text to Speech's billing system is pay-as-you-go for what you use. You will be charged for every 1,000,000 characters that are read aloud, including punctuation.
Speech Translation
Speech Translation converts speech into translated characters. It is similar to Speech to Text in terms of converting speech to text, but Speech Translation translates and outputs the converted text. Speech to Text also supports real-time translation, so you can translate what you're saying on the fly, just like YouTube video subtitles. Click here for the official website of Speech Translation
For example, it can be used for real-time video distribution and simultaneous interpretation of meetings.
Billing System
Speech Translation's billing system is pay-as-you-go. You are charged every hour for the time of the audio data you translate.
Speaker Recognition
Speaker Recognition can identify the speaker. You can identify who is speaking from the input audio data. You can register the audio of meeting participants in advance to identify who is speaking during the meeting or to find unregistered members. There is also a voice authentication function that authenticates the person speaking from the voice. Click here for the official website of Speaker Recognition
For example, voice authentication can recognize members using just their voice, or it can be combined with Speech to Text to automatically sort out who spoke to the meeting statement log.
Billing System
The billing system for Speaker Recognition is pay-as-you-go for what you use.
You are charged every 1,000 speaker identifications. In addition, identified speakers can register for free for up to 10,000 cases, and then they will be charged for every 1,000 cases.
Next time
In the next article, we will introduce the language field of Azure Cognitive Services.
Conclusion
In this article, we introduced the voice field of Azure Cognitive Services. The billing system for each service is pay-as-you-go, but if you store files for use in the cloud, you will be charged separately for storage. Colorkrew will provide support in various ways, from PoC to main development using Azure Cognitive Services. If you are interested in developing assistance with Azure Cognitive Services, please contact us at:
Cloud Utilization Support Service | Kuramane
