LessWrong audio: help us choose the new voice
We make AI narrations of LessWrong posts available via our audio player and podcast feeds:
LessWrong (Curated & Popular)
Curated posts and posts with 125+ karma.
https://pod.link/1630783021
LessWrong (30+ Karma)
Curated posts and posts with 30+ karma.
https://pod.link/1698192712
We’re thinking about changing our narrator’s voice.
There are three new voices on the shortlist. They’re all similarly good in terms of comprehension, emphasis, error rate, etc. They just sound different—like people do.[1]
We think they all sound similarly agreeable. But, thousands of listening hours are at stake, so we thought it’d be worth giving listeners an opportunity to vote—just in case there’s a strong collective preference.
Listen and vote
Please listen here:
And vote here:
It’ll take 1-10 minutes, depending on how much of the sample you decide to listen to.
Don’t overthink it—we’d just like to know if there’s a voice that you’d particularly love (or hate) to listen to.
We’ll collect votes until Monday December 16th. Thanks!
Other feedback?
We’re always keen for general feedback on the narration service. Please do share thoughts on the form, or in the comments on this post.
- ^
We periodically the various text-to-speech services, with a particular focus on ElevenLabs, Speechify, Azure, Amazon, BeyondWords, OpenAI, Murf.ai, Play.ht, Deepgram and a few open source libraries. The current shortlist was created based on factors like performance, reliability, customisation options and price. We won’t consider other voice options this winter, but if there are other voices you particularly like, please do let us know, and we’ll make sure they’re considered in future reviews. ↩︎
The results of the listener survey were equivocal. Listener preferences varied widely, with Ryan (our existing voice) and Echo coming out tied and slightly on top, but not with statistical significance.
Given that, we plan to stick with Ryan for now. Two considerations that influence this call, independently of the survey result:
There’s an habituation effect, such that switching costs for existing hardcore listeners are significant.
There’s some evidence that more expressive voices are less comprehensible at high listening speeds. Ryan is less expressive than the other voices we tested.
We’ll survey again—or perhaps just switch based on our judgement—when better models are released.