Technical Deep Dive: AI Accent Localization for Call Centers

voice-ai-newsletter.krisp.ai

4 pointsashvardanian3 comments

> An ideal dataset would consist of thousands of hours of speech where source accent utterance is mapped to each target accent utterance and aligned with it accurately.

To put it in terms of text translation, roughly how many sentences or words is this?

Pardon me, how many "tokens" ;-)

On average, an hour of speech contains about 9,000 to 15,000 words. This range accounts for different speaking speeds, which typically vary from 150 to 250 words per minute.

So this translates to tens of millions of words.

Reply on news.ycombinator.com