AmiVoice Communication Suite: Speech recognition solution for contact centers

Q: How is the recognition rate of voice recognition calculated?

The recognition rate is calculated by comparing the actual spoken content with the text of the speech recognition result. If every single character matches, the recognition rate is 100%, and if characters are inserted, deleted, or replaced, the recognition rate decreases. For example, if "Osewa ni narimasu" is recognized as "Osewa ni narimashita", out of 8 correct characters, one character is replaced from "su" to "shi" and one character is inserted as "ta", resulting in a recognition rate of (8-1-1)÷8 = 75%.

Q: Are there any methods or tips to improve the recognition rate of voice recognition?

The following methods can improve recognition accuracy: ■ Word Registration Accuracy can be improved by registering specialized terms and company jargon. Word registration can be easily done by the customer. ■ Individual Tuning of the Speech Recognition Engine Recognition accuracy can be improved by having our specialist staff train the speech recognition engine on the content of conversations at your contact center. ■ Operational Tips Recognition accuracy can be improved by paying attention to the following: Speak clearly and distinctly Set the recording volume appropriately Maintain an appropriate distance between the headset and your mouth

Q: Is a trial available?

Yes, we can conduct a proof-of-concept (POC) with a small number of users. We can also test speech recognition using pre-recorded audio data. Please contact us for details.

Q: Can it be implemented in a small call center?

Yes, it is possible. We will provide you with the best proposal tailored to your budget and scale. Please contact us for details.

Q: What makes it different from other companies' voice recognition systems?

AmiVoice Communication Suite employs a speech recognition engine specifically designed for contact center conversations. Leveraging extensive training data cultivated through implementations with over 600 companies, it can accurately transcribe call content. Furthermore, it supports individual tuning tailored to the customer's business operations and call characteristics, further enhancing speech recognition accuracy. In addition, it supports call content summarization, Q&A generation, and automatic customer information extraction using AI-generated content. Depending on security requirements, users can choose between a local LLM (Live Language Management) usable in a local environment and the latest cloud LLM, and hybrid operation combining both is also possible.

Q: What is the configuration of the voice recognition system?

We support both cloud-based and on-premises deployments. Furthermore, you can choose between real-time and batch processing. We can flexibly configure the system to suit your needs and environment. Please contact us for details.

Voice recognition x generative AI

Contact center operations
Dramatically more efficient

AmiVoice Communication Suite converts all calls into text in real time,
Automate response quality evaluation, sentiment analysis, summary generation, etc.
We provide thorough support for improving the efficiency of contact center operations.

ResourcesUnderstand it with numbers! (2-piece document set)

* Source: ITR "ITR Market View: Image and Voice Recognition Market 2025" Voice Recognition Market -
Sales share by vendor for contact center operations (FY2025 forecast)

600 companies have adopted the system.
Over 90,000 licenses!

Customer Harassment Detection & Countermeasures Package Banner

AmiVoice CQM Assist 3-month free trial campaign banner

Special page banner for local governments

All calls converted to text
Resolving Contact Center Issues

Reduce after-call work

All conversations between operators and customers are converted into text and generative AI is used to significantly reduce the work of creating response histories.

Improve customer service quality

The monitoring and chat functions allow for rapid escalation and thorough follow-up with operators.

Utilizing customer feedback
(VOCs)

By converting all calls into text, you can now use all of the customer feedback that you have previously missed for marketing analysis.

Compliance check
Efficiency

This enables the resolution of "said/not said" disputes and the rapid identification of risky calls, strengthening compliance.

Introduction Results

Reduce after-call work

Post-processing timeMore than 50% reduction,
Number of incoming calls8000 increase

Call centers in designated cities

The average post-processing time per call was reduced from 1 seconds to 56 seconds for inquiries and from 19 seconds to 35 seconds for main exchange operations, achieving an improvement rate of more than 16% compared to the previous manual input method. The number of calls received increased by approximately 50 per month, and the number of overflow calls decreased.

Increase sales

By visualizing sales talk
Improved conversion rate from 24% to 50%

SCSK Serviceware Corporation

Using voice recognition, we focused on the effect of the order of topics and the weighting of topics in sales talks on the success rate, and conducted a comparative analysis between high and low performers. We were able to easily and quickly extract the points of interest, and by implementing improvements, we were able to double the success rate.

Improve customer service quality

From the results of sentiment analysis
Talk script revision
Customer "delight" increased fourfold

SKY PerfecTV Customer Relations Co., Ltd.

Emotion analysis was used to compare the difference in customer emotions before and after improvements were made to the conversation script used to ask about cancellation reasons. After the improvements, customer happiness was four times greater, and the revised conversation script enabled responses that were more sympathetic to customer emotions.

Improving the efficiency of monitoring operations

In half the time
Approximately 150 times more monitoring

SB Payment Service Co., Ltd.

It has become possible to monitor 150 times as many calls in half the time it took before. In addition, the call content is automatically scored using an evaluation function based on HDI standards, which has made feedback clearer and led to quality improvements.

Operator support utilizing generational AI

Generative AI x Contact Center Optimal Solution

AI supports the entire business process, from during the call to after the call.

Multiple AI generators analyze call content, creating feedback checks and proposed answers. This allows operators to receive support at the optimal time without requiring any special operation. It covers the entire contact center workflow, from during the call to post-call processing.

AI assistant function

Real-time support from operators drastically reduces the burden on on-site staff.

The AI assistant covers the entire workflow, providing support tailored to the operator's situation.

Real-time support

AI agents automatically perform FAQ searches, information extraction, and knowledge presentation.

Even complex and advanced processes can be handled easily.

Multiple AI systems automatically collaborate to enable complex processing.

Automation of subsequent tasks

The extracted information is automatically linked to the CRM, reducing after-call work.

Flexible connection to various generative AI services

In addition to the secure local LLM (local generation AI) for contact centers, the latest generation AI services can also be connected. You can use them separately or in a hybrid configuration depending on your needs.

Local LLM

＆ High security

Customization

High accuracy

> Click here for AOI LLM implementation case studies

Cloud LLM

latest technology

High-performance and diverse services

Free prompt settings

Integration with generative AI services such as Azure OpenAI, Claude, and Gemini.

Abundant functions

Equipped with a wide range of functions, it thoroughly supports improving the efficiency of contact center operations.

Customer Harassment Detection and Countermeasures Package

Six functions of the Communication Suite have been customized for customer harassment detection and countermeasures.
A special package that combines a locally generated AI customer harassment information extraction model.
We provide strong support for customer harassment countermeasures, reflecting the know-how of 500 companies that have implemented our system.

Why AmiVoice is chosen in contact center environments

High recognition accuracy

The voice recognition engine has accumulated resources based on numerous implementation results, and converts voice to text with extremely high accuracy. Technical terms and abbreviations can also be converted smoothly and with high accuracy.

With a wide range of functions
easy-to-use interface

Not only does it have intuitive operability that allows you to efficiently access the information you need, but it also comes equipped with functions optimized for contact center operations.

API integration
Achieve more efficient operations

The API allows us to link with a variety of external systems and other companies' cloud services, enabling us to flexibly build systems that meet our customers' ideals.

Supports a variety of operational configurations

It can be operated in a variety of ways to suit the customer, such as cloud or on-premise, real-time or batch recognition, etc. It supports both server recognition and client recognition.

API integration with a wide range of external systems

It can easily be linked with various services, making your work more efficient.

Proven operational IP-PBX and cloud PBX services

Avaya Communication Manager Series
OKI CTstage series
NEC SV Series
NEC Aspire Series

Genesys PureConnect
Genesys PureCloud
Genesys Cloud
CT-e1

Fujitsu CL series
Fujitsu LEGEND V Series
Link BIZTEL
AmazonConnect

Bright Pattern Contact Center
Collabo @nyplace
Asterisk

We have experience with cases other than those listed above. Please contact us for details.

Reduce the burden and improve quality.
Why not start operating a next-generation contact center?

The effects are shown in numbers! 2-piece document set.

Resources

Feel free to ask us anything, even if it's just for information gathering!

Customer Reviews

BuySell Technologies Co., Ltd.

Reduced the time required to review recorded data by 80%, achieving improved quality and increased efficiency.

More details

Metropolitan Expressway Co., Ltd.

Customer Harassment Prevention Manual x Voice Recognition System to Improve Customer Center's On-site Capabilities

More details

Himeji

Himeji City, Hyogo Prefecture

Reduced overtime hours spent on recording telephone consultations by 66%

More details

Yamato Contact Service Co., Ltd.

Achieving VOC acquisition and improving customer service quality by introducing cloud services

More details

Persol Business Process Design Co., Ltd.

Automatic call evaluation function for monitoring improves the efficiency of evaluation work and response quality

More details

JAL Card Co., Ltd.

Real-time voice recognition dramatically improves the efficiency of call center operations

More details

Direct Solutions Co., Ltd.

The first use case in the insurance outbound telemarketing industry!
Greatly streamline the process of checking successful calls!

More details

[Column] What is a voice recognition service for call centers? Benefits and points to note when implementing it

FAQ

Q What is the recognition rate of the voice recognition?

Generally, when speaking politely and clearly, like a call center operator, the recognition rate is over 90-95% on average. On the other hand, when speaking casually, such as when speaking to a customer or on an internal line, the average rate is around 80-90%.

Q How is the recognition rate of voice recognition calculated?

It is calculated by comparing what was actually said with the text of the speech recognition results.
If every single character matches, the recognition rate will be 100%. If characters are inserted, deleted, or replaced, the recognition rate will decrease.
For example, if "Osewa ni namasu" is recognized as "Osewa ni shite natte," then for the eight correct characters, one character is replaced by "su" to "shi" and one character is inserted, resulting in a recognition rate of (8-1-1) ÷ 8 = 75%.

Q Can you recognize industry-specific terminology?

Yes, it is possible. Technical terms and proper nouns can be recognized by voice recognition by "word registration." You can easily register words yourself.

Q Does the app support speech recognition for dialects?

Differences in intonation can be recognized without any problems. However, for words and phrases specific to dialects, such as saying "thank you" instead of "thank you," it is necessary to register the words and train the speech recognition engine.

Q Are there any methods or tips to improve the recognition rate of voice recognition?

The following methods can be expected to improve recognition rates.

■ Word registration
Accuracy can be improved by registering technical terms and company terminology. Customers can easily register words themselves.

■ Individual tuning of the voice recognition engine
Our expert staff can improve the recognition rate by having the speech recognition engine learn the content of conversations at your contact center.

■ Operational improvements
You can improve the recognition rate by paying attention to the following:
Speak clearly and articulately
Set the recording volume appropriately
Maintain an appropriate distance between the headset and your mouth

Q Is a trial available?

Yes, we can do a proof of concept (POC) with a small number of seats. We can also provide you with pre-recorded voice data to test voice recognition. For details,Contact Us.

Q Can it be implemented in a small call center?

Yes, it is possible. We will make the best proposal based on your budget and scale. For details,Contact Us.

Q Is there a security issue with the call data handled by the generation AI?

Depending on your security requirements, you can choose between a local LLM that can be used in a local environment or a cloud-based generation AI. Local LLM does not require sending call data to an external cloud, and can be processed safely within the company's network. This reduces the risk of data leaks, even for highly confidential customer information and call content.

Q What makes it different from other companies' voice recognition systems?

AmiVoice Communication Suite uses a speech recognition engine specialized for contact center conversations. Utilizing the extensive learning data accumulated through implementation in over 600 companies, it can convert call content into text with high accuracy.
It also supports individual tuning to suit your business operations and call characteristics, further improving speech recognition accuracy. It also supports the use of generative AI to summarize call content, generate Q&A, and automatically extract customer information. Depending on your security requirements, you can choose between local LLM, which can be used in a local environment, or the latest cloud LLM, and even combine the two for hybrid operation.

Q What is the configuration of the voice recognition system?

It supports both cloud and on-premise systems. You can also choose between real-time processing and batch processing. It can be flexibly constructed to suit your needs and environment. For details,Contact Us.