This article is for companies building Voice AI Apps targeting the Contact Center. It outlines the key technical features, beyond accuracy, that are important while evaluating an OEM Speech-to-Text (STT) API. Usually, most analyses focus on the importance of accuracy and metrics like benchmarks of word error rates (WER). While accuracy is very important, there are other technical features that are equally important for contact center AI apps.
There are multiple use-cases for Voice AI Apps in the ContactCenter. Some of the common use cases are 1) AI Voicebot or Voice Agent 2) Real-time Agent Assist 3)Post Call Speech Analytics.
This article is focused on the third use-case which is Post-Call Speech Analytics. This use-case relies on batch STT APIs while the first two use-cases require streaming transcription. This Speech Analytics App helps the Quality Assurance and Agent-Performance management process. This article is intended for Product Managers and Engineering leads involved in building such AI Voice Apps that target the QA, Coaching and Agent Performance management process in the call center. Companies building such apps could include 1) CCaaS Vendors adding AI features, 2) Enterprise IT or Call Center BPO Digital organizations building an in-house Speech Analytics App 3) Call Center Voice AI Startups
Very often, call-center audio recordings are only available in mono. And even if the audio recording is in 2-channel/stereo, it could include multiple voices in a single channel. For example, the Agent channel can include IVR prompts and hold music recordings in addition to the Agent voice. Hence a very important criterion for an OEM Speech-to-Text vendor is that they provide accurate speaker diarization.
We would recommend doing a test of various speech-to-text vendors with a good sample set of mono audio files. Select files that are going to be used in production and calculate the Diarization Error Rate. Here is a useful link that outlines the technical aspects of understanding and measuring speaker diarization.
A very common requirement of Voice AI Apps is to redact PII – which stands for Personally Identifiable Information. PII Redaction is a post-processing step that a Speech-to-Text API vendor needs to perform. It involves accurate identification of entities like name, email address, phone number and mailing addresses and subsequent redaction both in text and audio. In addition, there are PCI – Payment Card Industry – specific named entities like Credit Card number, 3-digit PIN and expiry dates. Successful PII and PCI redaction requires post-processing algorithms to accurately identify a wide range of PII entities and cover a wide range of test scenarios. These test scenarios need to cover scenarios where there errors in user input and errors in speech recognition too.
There is another important capability related to PCI/PII redaction. Very often PII/PCI entities are present across multiple turns in a conversation between an Agent and Caller. It is important that the post-processing algorithm of the OEM Speech-to-Text vendor is able to process both channels simultaneously when looking for these named entities.
A Call Center audio recording could start off in one languageand then switch to another. The Speech-to-Text API should be able to detect language and then perform the transcription.
There will always be words that are not accurately transcribed by even the most accurate Speech-to-Text model. The API should include support for Hints or Keyword Boosting where words that are consistently misrecognized can get replaced by the correctly transcribed word. This is especially applicable for names of companies, products and industry specific terminology.
There are AI models that measure sentiment and emotion, and these models can be incorporated in the post-processing stage of transcription to enhance the Speech-to-Text API. Sentiment is extracted from the text of the transcript while Emotion is computed from the tone of the audio. A well-designed API should return Sentiment and Emotion throughout the interaction between the Agent and Caller. It should effectively compute the overall sentiment of the call by weighting the “ending sentiment” appropriately.
While measuring the quality of an Agent-Caller conversation, there are a few important audio-related metrics that are tracked in a call center. These include Talk-Listen Ratios, overtalk incidents and excessive silence and hold.
There are other LLM-powered features like computation of theQA Score and the summary of the conversation. However, these are features are builtby the developer of the AI Voice App by integrating the output of the Speech-to-TextAPI with the APIs offered by the LLM of the developer’s choice.
[UPDATE 1/23/22: After training on additional data, the Voicegain recognizer now achieves an average WER of 11.89% (an improvement of 0.35%) and a median WER of 10.82% (an improvement of 0.21%) on this benchmark.
Voicegain is now better than Google Enhanced on 44 files (previously 39).
Voicegain is now the most accurate recognizer on 12 of the files (previously 10).
We have additional data on which we will be training soon and will then provide a complete new set of results and comparison.]
It has been over 4 months since we published our last speech recognition accuracy benchmark. Back then the results were as follows (from most accurate to least): Amazon and Microsoft (close 2nd), then Google Enhanced and Voicegain (also close 4th) and then, far behind, IBM Watson and Google Standard.
Since then we have tweaked the architecture of our model and trained it on more data. This resulted in a further increase in the accuracy of our model. As far as the other recognizers are concerned, Microsoft improved the accuracy of their model the most, while the accuracy of others stayed more or less the same.
We have repeated the test using similar methodology as before: used 44 files from the Jason Kincaid data set and 20 files published by rev.ai and removed all files where the best recognizer could not achieve a Word Error Rate (WER) lower than 25%. Note: previously, we used 20% as the threshold, but this time we decided to keep more files with low accuracy to illustrate the differences on that type of files between recognizers.
Only three files were so difficult that none of the recognizers could achieve 25% WER. The two removed files were both radio phone interviews with bad quality of the recording.
As you can see in the results chart above, Voicegain is now better than Google Enhanced, both on average and median WER. Looking at the individual files the results also show the Voicegain accuracy is in most of the case better than Google:
Key observations about other results:
As you can see the field is very close and you get different results on different files (the average and median do not paint the whole picture). As always, we invite you to review our apps, sign-up and test our accuracy with your data.
When you have to select speech recognition/ASR software, there are other factors beyond out-of-the-box recognition accuracy. These factors are, for example:
1. Click here for instructions to access our live demo site.
2. If you are building a cool voice app and you are looking to test our APIs, click here to sign up for a developer account and receive $50 in free credits
3. If you want to take Voicegain as your own AI Transcription Assistant to meetings, click here.
You can find the complete code (minus the RASA logic - you will have to supply your own) at our github repository.
The setup allows you to call a phone number and then interact with a Voicebot that uses RASA as the dialog logic engine.
November 2021 Update: We do not recommend S3 and AWS Lambda for a production setup. A more up to date review of various options to build a Voice Bot is described here. You should consider replacing the functionality of S3 and AWS Lambda with a web server that is able to maintain state - like Node.js or Python Flask.
The sequence diagram is provided below. Basically, the sequence of operations is as follows:
The purpose of this blog post is to further elaborate on other posts in which we described various ways you can build a Voice Bot using Voicegain ASR/Speech-to-Text. We also plan to announce a new feature that will soon make Voice Bot development even easier.
Just a quick recap - what is a Voice Bot? A Voice Bot allows users to speak freely and naturally in response to questions asked by the Bot. It can extract multiple "intents" from what a customer says and can respond intelligently. By implementing Voice bots, customers can retire their legacy IVRs and also use a unified Bot platform to power both chatbots and Voice Bots.
It is important to note that Voicegain ASR/Speech-to-Text only provides the "mouth" and the "ear" of the Voice Bot. For building the bot logic and all the back-end integrations (i.e., the brains), a developer has to select a bot framework like Google Dialogflow, RASA, Kore.ai, Microsoft Azure Bot Service, or AWS Lex.
So here are ways you can build a Voice Bot.
This method is described in the blog post: How to build a Voicebot using Voicegain, Twilio, RASA, and AWS Lambda
The important thing to note is that the described setup of using AWS Lambda and S3 to handle the callbacks is for demo purpose only and not ideal for production deployment. The callback server has to be able to handle callbacks from Twilio and from Voicegain and pass information between the two. Because AWS Lambda is stateless the information is being passed in this example via S3 - it makes the end-to-end process slow because of the need for polling. That will not provide a fast response time for your Voice Bot.
For a production-ready setup we suggest replacing AWS Lambda and S3 with a proper web-server that is able to maintain session state - you could use Node.js or Python Flask for that.
This method is described in the blog post: Easy How-To: Build a Voicebot using Voicegain, RASA, and AWS Lambda
This is easier than the method described above. The Voicegain Telephony Bot API uses the Amazon Chime CPaas to provide the functionality otherwise provided by Twilio and this is internally integrated with Voicegain STT API. It uses callbacks, so it needs an intermediate web-service to handle the interaction with a bot platform, e.g. RASA. This web-service may be stateless because Telephone Bot API is capable of maintaining state information.
The example described in the above blog post uses SIP Trunks and phone numbers provided by Amazon Chime which is embedded as part of Voicegain Telephony Bot API. If you would rather retain your CPaaS/Telephony provider (e.g. SignalWire, Twilio, Telnyx, or Bandwidth.com) you can do that and connect to the Telephone Bot API using SIP INVITE. This is described in the blog post: SIP INVITE Voicegain from Twilio, SignalWire, Telnyx CPaaS
This method is described in the blog post: Voicegain announces integration with Audiocodes Voice AI connect.
AudioCodes VoiceAI Connect (VAIC) enables enterprises to connect a bot framework and speech services, such as text-to-speech (TTS) and speech-to-text (STT), to the enterprises’ voice and telephony channels to power Voice Bots, conversational IVRs and Agent Assist use-cases.
AudioCodes provides native integration with Bot Frameworks like Kore.ai, Google Dialogflow and Microsoft Bot Framework.
This setup allows you to directly specify a Voice Bot endpoint instead of specifying a generic http callback destination. The benefit of this is that you do not have to deal with having to provide the callback web-service. Notice that in this setup any back-end requests from your application logic to e.g. data services will now need to be done from the bot platform.
The bot platforms that we already support are RASA and Google Dialogflow. We are currently working on integrating with Microsoft Bot Framework. We hope to have this integration finished in time for the first release of Voicegain-Bot Platform integration. We also plan to very soon work on an integration with Kore.ai.
FreeSWITCH is a very capable telephony platform suitable for building various telephony applications. Some of those applications will rely speech-to-text conversion, for example: ACDs (automatic call distribution), IVRs, Voice-Bots, Real-Time Agent Assist, real-time conference call transcription, call monitoring, etc.
Voicegain Speech-to-Text platform can be used with FreeSWITCH in a variety of ways.
Voicegain STT platform has supported MRCP (Media Resource Control Protocol) for a long time now. Our ASR can be accessed using MRCP and we support both grammar-based recognition (e.g. GRXML) and large-vocabulary transcription. MRCP is a communication protocol designed to connect telephony based IVRs and Voice Bots with speech recognizers (ASR) and speech synthesizers (TTS).
FreeSWITCH can interact with MRCP based recognizers using the included mod_unimrcp module. Voicegain STT has been tested with mod_unimrcp and interfaces with it without problems. You can learn more about using Voicegain STT via mod_unimrcp in this blog post.
Voicegain supports MRCP both in the Cloud and on the Edge (on-prem). We will soon be releasing in OpenSource a recognizer plugin for unimrcp server that will give you even more options in deploying FreeSWITCH with Voicegain and MRCP.
Voicegain provides a Telephony Bot API which is a callback API - similar in style to Twilio TwiML. You can place a call to Voicegain endpoint either using a phone number obtained from Voicegain or using a SIP endpoint unique to your Voicegain application. When a call arrives you will get a web callback and the response you will provide will determine actions that the Voicegain platform will perform, like e.g. play a prompt, recognize speech, detect DTMF, etc.
You can learn more about this API from the following blog posts:
If you have a FreeSWITCH application and you would like to recognize spoken speech you can bridge into Voicegain SIP endpoint and in a callback specify a prompt and the type of speech capture (grammar-based or large vocabulary). Once the recognition finishes you will get a callback and then you can either issue a disconnect command which will transfer call flow back to your Freeswitch app, or you can continue with additional questions and recognitions on Voicegain platform as needed.
Below is an example of a simple interaction with 4 participants:
This is still not Generally Available - please contact us if you are interested in testing.
mod_voicegain will give you capabilities similar to using mod_unimrcp with Voicegain but without the whole overhead of using an MRCP protocol - mod_voicegain talks directly to Voicegain ASR.
mod_voicegain taps into the FreeSWITCH inbound audio stream and sends the audio data to Voicegain ASR in the Cloud or on the Edge. Voicegain ASR processes the audio according to the invocation parameters specified in the data argument. It then communicates the result of transcription or recognition in an Event.
mod_voicegain installs on FreeSWITCH as an app and can be invoked as a such, e.g.:
or from LUA script:
Results will always be returned as a FreeSWITCH event but it is also possible to get the results in a callback to the url specified in callback.uri
The FreeSWITCH event will be of custom type (Event-Name: CUSTOM) and Event-Subclass will be "voicegain_asr_update". The relevant payload will be in the "ASR-Response" field formatted as JSON.
You can read more about mod_voicegain is this Knowledge Base article.
mod_vg_tap has been developed with applications like Real-Time Agent Assist in mind. These apps need access to the audio stream from a FreeSWITCH call but do not otherwise need to interact with FreeSWITCH (unlike IVR and Voice-Bots).
mod_vg_tap installs as an app and has simple commands to start/stop streaming to Voicegain Speech-to-Text engine.
The start command can specify the following destinations:
The results from transcription are generally not returned to a FreeSWITCH app but will be delivered to the destination specified when starting speech-to-text session - the results can be delivered via websocket, polling, or callback.
If you want more information about any of these methods of integrating Voicegain with FreeSWITCH, please email us at support@voicegain.ai.
Dallas, Texas - October 26, 2021: OnviSource, a leading provider of intelligent automation solutions for workforce optimization, contact center operation analytics and automation, customer experience management, and business process automation, announced today a strategic partnership with Voicegain, an innovative Speech-to-Text/ASR company. OnviSource has integrated Voicegain’s deep learning-based speech-to-text platform into its Intellecta™ multichannel analytics solution which utilizes speech-to-text and natural language understanding to analyze customer interactions and audio-based content to discover actionable knowledge and extract business insights.
OnviSource will leverage the Voicegain platform to serve its growing enterprise client base from various industries such as nationwide wireless service providers, banking, financial services, utilities, insurance and others.
“We are pleased to announce this partnership with Voicegain as their AI-driven ASR further augments our AI-driven intelligent automation solutions and our hyper-automation platform that offers integrated AI, conversational AI, RPA, BPA and analytics,” said Ray Naeini, Chairman and CEO of OnviSource. “Our partnership will allow both companies to jointly develop highly sophisticated and customized AI models for various applications and industries in order to deliver unmatched accuracy and performance.”
To achieve high performance, OnviSource deployed the Voicegain ASR Engine on servers with NVIDIA GPUs in its data center. This architecture is referred to as an Edge deployment. While Voicegain also offers a multi-tenant cloud solution, an Edge deployment architecture has two important benefits for OnviSource.
The first major benefit is that it allows OnviSource to meet strict customer contractual commitments related to data privacy, security and control. The second benefit is that it delivers approximately a 75% reduction in costs for OnviSource compared to usage-based pricing models provided by other providers, empowering OnviSource to offer its feature-rich solutions at highly affordable and flexible prices.
“We are excited to be selected by OnviSource for its call center and enterprise speech analytics products. This decision validates the ‘3As’ on which Voicegain differentiates itself in the ASR market – Accuracy, Affordability and Accessibility,” said Arun Santhebennur, Co-founder & CEO of Voicegain. “Our joint product enhancements will deliver highly accurate Speech-to-Text models for complex business applications.”
Selection of the Voicegain product by OnviSource was based on comprehensive trials and pilot programs related to accuracy, performance and applicability of Voicegain’s product, combined with detailed comparative analysis with other products in the market.
Additionally, the Voicegain product offers simplicity in deployment and usage as the entire platform is deployed on a Kubernetes cluster. Its Edge deployment offers a simple script to download and deploy all the packages and dependencies on any server with NVIDIA GPUs.
About OnviSource
For more than a decade, OnviSource has enabled several hundred small-to-large companies across a broad range of industries to cost-effectively manage, automate and improve their customer experience and business processes by offering advanced solutions in multichannel data and media capture, unification, analysis, decision making and automation for their entire enterprise, including their contact centers, back offices and IT organizations.
OnviSource ia.Enterprise Intelligently Automated (IA) solutions offer Workforce Optimization and Workforce Management (WFO/WFM), inclusive Teleservice Customer Engagement Management, Multichannel Customer Engagement Analytics, intelligently automated Customer Survey, Process Automation through Robotic Process Automation (RPA) and Intelligent Process Automation (IPA) and Intelligent Virtual Agent (IVA). The Company delivers its solutions as software products, cloud or Software-as-a-Service (SaaS), managed services, or any combination. OnviSource’s special Advantage Platinum program assures that solutions work for customers’ specific needs by offering a series of customer assistance programs with no obligations. These programs include consultation, proof-of-concept and hands-on operation assistance. OnviSource is headquartered in Plano, Texas (North Dallas area), with an additional operation center in Oklahoma.
About Voicegain
Voicegain is a deep neural network-based Speech-to-Text platform that is focused on developers of voice applications. Voicegain offers a full suite of APIs, SDKs and SaaS apps on top of its platform to automate and analyze voice-based interactions in contact centers, sales and meetings. To learn more, visit Voicegain.ai or create a free account to get started.
Press Contact:Voicegain: Arun Santhebennur, CEO
OnviSource: Deborah Cromwell, Marketing Manager
deborah.cromwell@onvisource.com
One of the previous blog posts described a Voice Bot built using Twilio, Voicegain, RASA, and AWS Lambda. Twilio was used for telephony (phone numbers, SIP Trunking, TwiML for call control) Voicegain provided the ASR/speech recognition, while AWS Lambda was coordinating the actions. The setup works but is involved. The need to pass the speech recognition results via S3 (as Lambda is stateless and does not have memory between function calls) may occasionally cause delays in requests and responses.
Voicegain now integrates with Amazon Chime Voice Connector to offer a pay as you go SIP Trunking service directly from the Voicegain web console. You can also purchase phone numbers and receive inbound calls. Support for making outbound Speech IVR calls is in the works.
Of course, we continue to support developer that use Twilio and SignalWire using simple SIP INVITE - this blog describes how.
The sequence diagram is provided below. It is very simple. Basically, the sequence of operations is as follows:
The sample code for the Lambda function (in python and node.js versions) is available on our github.
1. Click here for instructions to access our live demo site.
2. If you are building a cool voice app and you are looking to test our APIs, click hereto sign up for a developer account and receive $50 in free credits
Donec sagittis sagittis ex, nec consequat sapien fermentum ut. Sed eget varius mauris. Etiam sed mi erat. Duis at porta metus, ac luctus neque.
Read more →
Donec sagittis sagittis ex, nec consequat sapien fermentum ut. Sed eget varius mauris. Etiam sed mi erat. Duis at porta metus, ac luctus neque.
Read more →Donec sagittis sagittis ex, nec consequat sapien fermentum ut. Sed eget varius mauris. Etiam sed mi erat. Duis at porta metus, ac luctus neque.
Read more →Donec sagittis sagittis ex, nec consequat sapien fermentum ut. Sed eget varius mauris. Etiam sed mi erat. Duis at porta metus, ac luctus neque.
Read more →Donec sagittis sagittis ex, nec consequat sapien fermentum ut. Sed eget varius mauris. Etiam sed mi erat. Duis at porta metus, ac luctus neque.
Read more →Donec sagittis sagittis ex, nec consequat sapien fermentum ut. Sed eget varius mauris. Etiam sed mi erat. Duis at porta metus, ac luctus neque.
Read more →Interested in customizing the ASR or deploying Voicegain on your infrastructure?
Voicegain helps developers build awesome voice enabled apps by providing them with the most accurate, affordable and accessible Speech-to-Text platform.
'%3e%3cg id='Final-Copy-2_2_' transform='translate(1275.000000, 200.000000)'%3e%3cpath class='st0' d='M7.4,12.8h6.8l3.1-11.6H7.4C4.2,1.2,1.6,3.8,1.6,7S4.2,12.8,7.4,12.8z'/%3e%3c/g%3e%3c/g%3e%3c/g%3e%3cg id='final---dec.11-2020'%3e%3cg id='_x30_208-our-toggle' transform='translate(-1275.000000, -200.000000)'%3e%3cg id='Final-Copy-2' transform='translate(1275.000000, 200.000000)'%3e%3cpath class='st1' d='M22.6,0H7.4c-3.9,0-7,3.1-7,7s3.1,7,7,7h15.2c3.9,0,7-3.1,7-7S26.4,0,22.6,0z M1.6,7c0-3.2,2.6-5.8,5.8-5.8 h9.9l-3.1,11.6H7.4C4.2,12.8,1.6,10.2,1.6,7z'/%3e%3cpath id='x' class='st2' d='M24.6,4c0.2,0.2,0.2,0.6,0,0.8l0,0L22.5,7l2.2,2.2c0.2,0.2,0.2,0.6,0,0.8c-0.2,0.2-0.6,0.2-0.8,0 l0,0l-2.2-2.2L19.5,10c-0.2,0.2-0.6,0.2-0.8,0c-0.2-0.2-0.2-0.6,0-0.8l0,0L20.8,7l-2.2-2.2c-0.2-0.2-0.2-0.6,0-0.8 c0.2-0.2,0.6-0.2,0.8,0l0,0l2.2,2.2L23.8,4C24,3.8,24.4,3.8,24.6,4z'/%3e%3cpath id='y' class='st3' d='M12.7,4.1c0.2,0.2,0.3,0.6,0.1,0.8l0,0L8.6,9.8C8.5,9.9,8.4,10,8.3,10c-0.2,0.1-0.5,0.1-0.7-0.1l0,0 L5.4,7.7c-0.2-0.2-0.2-0.6,0-0.8c0.2-0.2,0.6-0.2,0.8,0l0,0L8,8.6l3.8-4.5C12,3.9,12.4,3.9,12.7,4.1z'/%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/svg%3e)
Your Privacy Choices