Sakhanda Wire
NVDA $153.63 -1.61% MSFT $506.84 -2.15% GOOGL $280.27 -1.22% META $619.77 -1.06% AMZN $256.65 -0.43%
← Back to the news

Gradium Introduces Voice Design for Creating Synthetic Voices from Text

Gradium, a Paris-based voice AI company spun out of the Kyutai research lab, has launched Voice Design. The service turns a written description into up to five new synthetic voice candidates, without reference audio or a specific speaker. Descriptions can specify attributes such as age, accent, pitch, pace, energy, timbre and intended use, and may contain 1 to 500 characters in English, French, Spanish, Portuguese or German.

Voice Design is available through Gradium’s API and Studio at no additional cost on all plans, including the free tier. Candidates are usually generated in 3 to 5 seconds and can be tested through the regular Text-to-Speech endpoint. A selected candidate can be converted into a production voice, which then works with REST, WebSocket and Speech-to-Speech services. Unconverted candidates expire after 30 days; the free tier includes five custom voice slots, while paid plans include 1,000.

Gradium says its blind listening test across five languages produced a 72.6% win rate in 7,627 comparisons, ahead of the systems tested. The company also reported first place in each language and an 83.4% score on the English split of InstructTTSEval. The evaluations were designed and conducted by the vendor.

Written by Sakhanda Wire AI from reporting by MarkTechPost. Facts, quotes and full context are in the original article below.

Originally published by MarkTechPost on

Read the original on MarkTechPost ↗

The reporting and the image belong to MarkTechPost and are credited and linked above; the text on this page is Sakhanda Wire's own summary.

← Back to the news

More stories

All the latest news