Veritone Adds Advanced Capabilities, Voices, and Languages to Veritone Voice
Veritone has unveiled its next-generation Veritone Voice with new application capabilities, more than new 70 AI voices, hundreds of new stock voices, and more than 15 new languages, including Albanian, Arabic, Mongolian, and Nepali.
The recent acquisition of synthetic voice pioneer VocaliD made these additions possible.
"As the market for AI voice solutions grows, it is becoming increasingly important that the generated voices sound and feel hyper-realistic. Many elements need to be taken into account and managed, including the tone of these voices, such as being happy, somber, or promotional. In today's digital world, monotone digital-sounding voices simply won't cut it," said Sean King, senior vice president and general manager of commercial enterprise at Veritone, in a statement. "Addressing these needs is at the heart of our latest version of Veritone Voice, which empowers the voice to be articulate, genuine, and nuanced for the content and audience. Whether the voices are used for broadcast-quality productions, the metaverse, or other applications, we believe that content creators and brands will appreciate the authenticity our Veritone Voice solutions offer as they move their projects forward."
In addition to the new voice-over voices, Veritone added other enhancements, such as Lexicon, that bring more custom implementation and expedite and streamline workflow processes. Lexicon enables customers to use their own dictionary of language so voice outputs recognize their custom terminology.
In addition, Veritone's Speech-to-Speech offering is now fully integrated with Respeecher, a voice conversion and voice cloning engine. STS clip creation is now available within Veritone Voice. Once the voice model is generated, users can create STS clips on their own.
Other advanced features include built-in intonation and upgraded editing. Intonation is a shortcut for pitch adjustment that allows the user to drag points on a timeline where they want the pitch to go up or down. With editing, users can now edit text post-clip creation with built-in auto save for frequent users to pick up where they left off.
Veritone Voice offers a suite of integrated voice features for creation, management, workflows, monetization, and licensing with compliance and protection. Built on Veritone aiWARE, Veritone Voice offers voice engines that combine with other cognitive capabilities, such as translation, sentiment analysis, content classification, and more, to create content at scale. To protect a creator's synthetic voice, Veritone includes inaudible watermarks, traceability, licensing protocols, and tools to protect and place claims against unauthorized monetization of content on social platforms. The Veritone Voice solution supports both speech-to-speech and text-to-speech.
Veritone's synthetic speech technology will power Stats Perform's live sports commentary.