# Voice recognition in base-64 encoded

**URL:** <https://community.brightpattern.com/t/voice-recognition-in-base-64-encoded/233>\
**Category:** API and Integrations\
**Created:** [December 9, 2020, 6:15am UTC](https://community.brightpattern.com/t/voice-recognition-in-base-64-encoded/233 "2020-12-09T06:15:50Z")\
**Posts on this page:** 6\
**Page:** 1

<div class="post-metadata">

**Author:** ![Salvador.Mendoza](https://sea2.discourse-cdn.com/flex020/user_avatar/community.brightpattern.com/salvador.mendoza/32/106_2.png) [@Salvador.Mendoza](https://community.brightpattern.com/u/Salvador.Mendoza)\
**Post date:** [December 9, 2020, 6:15am UTC](https://community.brightpattern.com/t/voice-recognition-in-base-64-encoded/233/1 "2020-12-09T06:15:50Z")

</div>

Hello!! I have a customer that needs voice recognition. Does anyone know if there is an API or SDK in the Contact Center Software or if there is a way to do Voice Recognition in base 64 encoded format?

---

<div class="post-metadata">

**Author:** ![sergey.menshikov](https://sea2.discourse-cdn.com/flex020/user_avatar/community.brightpattern.com/sergey.menshikov/32/6_2.png) [@sergey.menshikov](https://community.brightpattern.com/u/sergey.menshikov)\
**Post date:** [December 9, 2020, 5:06pm UTC](https://community.brightpattern.com/t/voice-recognition-in-base-64-encoded/233/2 "2020-12-09T17:06:30Z")

</div>

Salvador,

If your case is conversational IVR, we would recommend configure integration for Google Speech To Text and use Play-Listen block. This works well with a bot, e.g. Watson.

> **[5.3:Tutorials-for-admins/Voice/ConversationalIVRConfiguration - Bright...](https://help.brightpattern.com/5.3:Tutorials-for-admins/Voice/ConversationalIVRConfiguration)**

let me know if your customer’s use case is different.

Sergey

---

<div class="post-metadata">

**Author:** ![Salvador.Mendoza](https://sea2.discourse-cdn.com/flex020/user_avatar/community.brightpattern.com/salvador.mendoza/32/106_2.png) [@Salvador.Mendoza](https://community.brightpattern.com/u/Salvador.Mendoza)\
**Post date:** [December 9, 2020, 6:08pm UTC](https://community.brightpattern.com/t/voice-recognition-in-base-64-encoded/233/3 "2020-12-09T18:08:46Z")

</div>

Thanks a los Sergey! In this case… we need to identify the caller by voice, saying a particular phrase.

Salvador

---

<div class="post-metadata">

**Author:** ![sergey.menshikov](https://sea2.discourse-cdn.com/flex020/user_avatar/community.brightpattern.com/sergey.menshikov/32/6_2.png) [@sergey.menshikov](https://community.brightpattern.com/u/sergey.menshikov)\
**Post date:** [December 10, 2020, 6:11pm UTC](https://community.brightpattern.com/t/voice-recognition-in-base-64-encoded/233/4 "2020-12-10T18:11:22Z")

</div>

In that case, I would recommend using VoiceIT, you would just need a base64-to-file-upload converter function, e.g. in PHP or AWS Lambda. Barebone PHP examples are here [https://www.dropbox.com/s/kpc2iwsrosop8r1/voiceit.zip?dl=0](https://www.dropbox.com/s/kpc2iwsrosop8r1/voiceit.zip?dl=0)

---

<div class="post-metadata">

**Author:** ![Salvador.Mendoza](https://sea2.discourse-cdn.com/flex020/user_avatar/community.brightpattern.com/salvador.mendoza/32/106_2.png) [@Salvador.Mendoza](https://community.brightpattern.com/u/Salvador.Mendoza)\
**Post date:** [December 10, 2020, 6:57pm UTC](https://community.brightpattern.com/t/voice-recognition-in-base-64-encoded/233/5 "2020-12-10T18:57:03Z")

</div>

Great!! I will try with that options and let you know the results.

Salvador Mendoza

---

<div class="post-metadata">

**Author:** ![Salvador.Mendoza](https://sea2.discourse-cdn.com/flex020/user_avatar/community.brightpattern.com/salvador.mendoza/32/106_2.png) [@Salvador.Mendoza](https://community.brightpattern.com/u/Salvador.Mendoza)\
**Post date:** [December 17, 2020, 5:39pm UTC](https://community.brightpattern.com/t/voice-recognition-in-base-64-encoded/233/6 "2020-12-17T17:39:47Z")

</div>

It’s already working! It takes a while to recognize the voice, I don’t know why but we have already achieved the integration. We are going to do the demo with the client to see how it goes us.