Transcribe speech to text in real time from phone calls

Speech-to-text (STT) has been around since the 1950s. But in the past few years, it’s become a crucial tool in building conversational AI, enabling your applications to understand and respond to spoken words.

Using STT, you can use every phone call to unlock actionable insights with just a click. By leveraging the power of conversational AI and STT, every call has the potential to enhance customer understanding, improve service, drive innovation, and boost retention.

Telnyx Programmable Voice API and TeXML offer seamless gateways to transform voice interactions into text, making every word count. Whether you're a developer seeking to integrate cutting-edge tech or a business owner aiming for powerful customer insights, our upcoming demo will navigate you through the simplicity and power of real-time speech recognition. Equipped with these tools, you can enhance your apps for customer support, voice commands, or any other service that requires voice interaction with speech-to-text capabilities.

Use speech-to-text to create more interactive and user-friendly applications

Real-time speech-to-text transcription allows you to transcribe text on a live phone call. And with Telnyx’s tools, you can create those transcriptions automatically, with nothing more than an API command or a few lines of XML.

Our next-generation, developer-friendly Voice API enables advanced conversational AI applications through a wide range of features or the best conversational AI interactions, including:

• Programmable on-demand call recording
• Real-time media forking
• HD voice
• Noise suppression

Customers like Replicant and echowin are already using Telnyx to build the next generation of intelligent voice applications.

How to get started with speech-to-text

There are two ways to get started with speech-to-text using Telnyx:

1. Our Voice API
2. The dynamic, low-code markup language TeXML.

We’ve demonstrated speech transcription using both Telnyx Voice API and TeXML in the videos below. We’ll walk you through the whole process of transcribing audio from a live call—all in under five minutes.

Before you can follow along with the steps in the video, you’ll need to create a Telnyx account. From there, purchase a phone number, and set up your developer environment. For help, refer to our quickstart guide. Then, follow the steps in the video below to start transcribing in just a few minutes.

Getting started with speech-to-text using Voice API

Once you've created your account and gotten set up with a phone number and Outbound Voice Profile associated with your application, you'll need to grab an API key. Simply head to the API Keys section of the portal and create one.

From there, our video demo walks through everything you need to start transcribing phone calls. Our API Reference has ready-to-copy code for the Dial and Transcription Start commands used in the video.

How to Transcribe Speech to Text in Real-Time from a Phone Call from Telnyx on Vimeo.

Getting started with speech-to-text using TeXML

This video goes through the following steps:

1. Creating a new TeXML Bin and pasting in the transcription script from our TeXML Translator section of dev docs.
2. Creating a new TeXML application.
3. Assigning a phone number to your new TeXML application.
4. Calling the phone number and watching your transcription enterd your terminal in real-time.

Getting started with speech-to-text using TeXML from Telnyx on Vimeo.

How much does real-time on-call transcription cost?

Our default Google Speech-to-Text Transcription is priced at $0.05 / minute and billed in 15-second increments. However, we recently introduced our new highly-accurate in-house speech to text transcription engine, which starts at $0.025 per minute.

Use the Telnyx Voice API or TeXML to transcribe phone calls in real time

Real-time Speech-to-text enhances the user experience, makes tools more accessible to those with disabilities, resulting in improved operational efficiency and natural conversational AI interactions.

Telnyx's Voice API and TeXML are designed for ease of use, offering straightforward, intuitive interfaces that allow developers to integrate voice capabilities into their applications with minimal hassle, reducing the time and resources required for implementation, enabling developers to focus on creating innovative solutions rather than grappling with complex integration processes. In fact, our Mission Control Portal was awarded “Most Implementable CPaaS Platform by G2! This user-friendly approach makes getting started with integrating voice features like speech transcription to apps easy.