Dark Hacker News
new
|
best
|
ask
|
show
|
jobs
Grok Voice Transcribe 2.0
(x.ai)
22 points
by
vertigoruntime
4 hours ago
| 7 comments
Grok Voice Transcribe 2.0 | Dark Hacker News
nomel
3 hours ago
|
next
[−]
> Word-level timestamps. Each word has precise start/end times and confidence scores.
Neat. This is missing from many models.
throwaway89201
2 hours ago
|
parent
|
next
[−]
Mistral STT offers it, which is an open model.
xnx
3 hours ago
|
next
[−]
$0.10/hour is a good price vs $0.30/hour for Gemini Transcribe 3.5.
exizt88
3 hours ago
|
next
[−]
I don't understand how these benchmarks get things so wrong. Nova-3 is by far the best model on the market in WER. Azure's STT is atrocious. And yet every benchmark rates Azure's STT higher than Nova-3's.
mistic92
3 hours ago
|
parent
|
next
[−]
Mai 2 transcribe is giving me better results than nova
mistic92
3 hours ago
|
next
[−]
If only they had ZDR
bigyabai
4 hours ago
|
next
[−]
No weights, sadly. Bit of a nothingburger release.