Article·Announcements·May 5, 2023

Now Available: Deepgram Speech Recognition for Twilio Programmable Voice

Katie Byrne
By Katie Byrne
PublishedMay 5, 2023
UpdatedJun 13, 2024

Developers love Twilio, and developers love Deepgram. Now they'll love that it’s even easier to unlock the value of post-call recordings through the new Deepgram AI Speech Recognition Add-On in the Twilio Marketplace.

What can developers do with post-call transcription?

  • Add sentiment analysis to every recorded call

  • Identify topic trends by contact center location

  • Analyze discussion points from recruiting and/or sales calls for coaching purposes

  • And much more! Check out some examples built with Deepgram.

Post-Call Transcription Has Extended beyond Contact Center into the Enterprise

For years contact centers have recorded their calls for training and quality assurance purposes, however it wasn’t until the recent past that it was not only possible but economically viable to transcribe 100% of those calls. Call recording is now extending beyond the contact center and support to other departments within an organization such as sales, marketing, and recruiting. Regardless of department, each voice initiative is underway with the goal of either increasing their employee productivity, reducing cost or finding new revenue streams to help them gain a competitive advantage. Call recordings, more than many other touch points with our customers, are rich with information. Beyond understanding correctly what words were said, voice is rich with additional metadata such as the sentiment about the topic at hand.

Introducing Deepgram AI Speech Recognition for Twilio

Deepgram has invested in our partnership and integration with Twilio to make it even easier for developers to use the services to build amazing experiences without the hassle of connecting the APIs together.

With the new Deepgram AI Speech Recognition Add-on Twilio users can:

  • Get their first transcript in 10 minutes or less.

  • Transcribe 1 hour of audio in as quickly as 15s – thats anywhere from 95% to 99% faster depending which Speech Recognition add-on you used previously

  • Leverage the industry's highest accuracy. Many Developers see over 90% accuracy out of the box with Deepgram depending on their use case.

  • Easily add features such as punctuation, profanity filters, redaction, speaker diarization and more to their transcripts.

Get Started in Twilio Console

From the Twilio console, Developers can easily add Deepgram to their Programmable Voice project (sign up here). Once they have selected their preferred model tier, Base or Enhanced, they are presented with configuration options. You don’t need to create a separate Deepgram account on our website or any other steps. You simply select the speech recognition and understanding settings you’d like from the Twilio console, and away you go.

Tons of Features available in the Deepgram Speech Recognition Add-On for Twilio

Twilio users can leverage these features to format and/or classify their transcriptions:

  • Punctuation

  • Profanity Filter

  • Redaction

  • Diarization

  • Named Entity Recognition (NER)

  • Multichannel

  • Numerals

  • Callback

  • Paragraphs

  • Utterances

  • Utterance Split

  • Interim Results

To learn more about the Deepgram Add-on for Twilio, please visit the Twilio Marketplace and checkout the Deepgram listing.

If you have any feedback about this post, or anything else around Deepgram, we'd love to hear from you. Please let us know in our GitHub discussions .

Unlock language AI at scale with an API call.

Get conversational intelligence with transcription and understanding on the world's best speech AI platform.