Live data from Hacker News

Lyra audio codec enables high-quality voice calls at 3 kbps bitrate

cnx-software.com

11–20 of 207 posts

Re: Lyra audio codec enables high-quality voice calls at 3 kbps bitrate

#13

When Google's announcement [1] was posted a few days ago, I listened to their samples and heard an odd effect in the "chocolate bread" sample (the video chat example) [1], which is not mirrored in this article. On that sample, I felt [2] that the Lyra version exaggerates the pronunciation of the phrase 'with chocolate' in a way that meaningfully differs from the speaker's original. It weakens the voiced 'th' to nothi…

Are the speech models sufficiently generic across all languages?

Re: Lyra audio codec enables high-quality voice calls at 3 kbps bitrate

#14
post #3

If the demos are actually representative, it does seem impressive. Could save a lot of bandwidth for VoiP if it replaced 8kb/s G729.

Isn‘t VoIP at such low data rates already dominated by the overhead of UDP, IP and whatever lower layer? Multiplexing it with a low-bandwidth video stream would be possible, though.

I was thinking this could be most relevant for something like digital wireless transmissions.

Re: Lyra audio codec enables high-quality voice calls at 3 kbps bitrate

#16

When Google's announcement [1] was posted a few days ago, I listened to their samples and heard an odd effect in the "chocolate bread" sample (the video chat example) [1], which is not mirrored in this article. On that sample, I felt [2] that the Lyra version exaggerates the pronunciation of the phrase 'with chocolate' in a way that meaningfully differs from the speaker's original. It weakens the voiced 'th' to nothi…

I read your comment before I watched that video and I can't stop laughing. It sounds ridiculous!

Re: Lyra audio codec enables high-quality voice calls at 3 kbps bitrate

#17
post #6
post #2

Maybe in the future, all we need is a speech example, some AI and the continious transmission of text for low data voice transmission?

I think this will be the rough direction, but not exactly text, rather some other efficient, machine-readable embedding of speech that is also able to carry tone and rhythm effectively and pronounciation accurately and unambiguously.

Basically yes. "Features, or distinctive speech attributes, are extracted from speech every 40ms and are then compressed for transmission. The features themselves are log mel spectrograms, a list of numbers representing the speech energy in different frequency bands, which have traditionally been used for their perceptual relevance because they are modeled after human auditory response."

Re: Lyra audio codec enables high-quality voice calls at 3 kbps bitrate

#18
post #4

Is it any good for languages other than English?

"As with any ML based system, the model must be trained to make sure that it works for everyone. We’ve trained Lyra with thousands of hours of audio with speakers in over 70 languages using open-source audio libraries and then verifying the audio quality with expert and crowdsourced listeners. ... Lyra trains on a wide dataset, including speakers in a myriad of languages, to make sure the codec is robust to any situation it might encounter."

Re: Lyra audio codec enables high-quality voice calls at 3 kbps bitrate

#19
post #14
post #3

If the demos are actually representative, it does seem impressive. Could save a lot of bandwidth for VoiP if it replaced 8kb/s G729.

Isn‘t VoIP at such low data rates already dominated by the overhead of UDP, IP and whatever lower layer? Multiplexing it with a low-bandwidth video stream would be possible, though. I was thinking this could be most relevant for something like digital wireless transmissions.

To say the least, yeah. At 3kbps and 20ms framing, it's only 7.5 bytes of payload per frame.

RTP, UDP, IP, and Ethernet overhead are what - 60-ish bytes?

Post reply on HN