The origins of VoiceXML began in 1995 as an XML-based dialog design language intended to simplify the speech recognition application development process within an AT&T project called Phone Markup Language (PML). 

As AT&T reorganized, teams at AT&T, Lucent and Motorola continued working on their own PML-like languages. 

In 1998, W3C hosted a conference on voice browsers. By this time, AT&T and Lucent had different variants of their original PML, while Motorola had developed VoxML, and IBM was developing its own SpeechML. Many other attendees at the conference were also developing similar languages. The VoiceXML Forum was then formed by AT&T, IBM, Lucent, and Motorola to define a standard dialog design language that developers could use to build conversational applications. They chose XML as the basis for this.

In 2000, the VoiceXML Forum released VoiceXML 1.0 to the public. Shortly thereafter, VoiceXML 1.0 was submitted to the W3C as the basis for the creation of a new international standard. 

VoiceXML 2.0 is the result of this work based on input from W3C Member companies, other W3C Working Groups, and the public. VoiceXML 2.0 reached the final "Recommendation" stage in March 2004.

Major design goals of VXML were: 

Minimizes client/server interactions by specifying multiple interactions per document.

Shields application authors from low-level, and platform-specific details.

Separates user interaction code (in VoiceXML) from service logic (e.g. CGI scripts).

Promotes service portability across implementation platforms. VoiceXML is a common language for content providers, tool providers, and platform providers.

VoiceXML 2.1 added a relatively small set of additional features to VoiceXML 2.0, based on feedback from implementations of the 2.0 standard. It is backward compatible with VoiceXML 2.0 and reached W3C Recommendation status in June 2007.

VoiceXML 3.0 will be the next major release of VoiceXML, with new major features. It includes a new XML statechart description language called SCXML.