Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
snakers41
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
7 ms
·
31.
▲
by
snakers41
5y ago
Check out silero-models, available in colab with 10 speakers, running 3-4 RTS on one CPU core: - https://github.com/snakers4/silero-models#text-to-speech - https://habr.com/ru/post/549482
32.
▲
by
snakers41
5y ago
You should also totally check out silero-models, which are also available in colab with 10 speakers: - https://github.com/snakers4/silero-models#text-to-speech - https://habr.com/ru/post/54948
33.
▲
by
snakers41
5y ago
You should also totally check out silero-models, which are also available in colab with 10 speakers: - https://github.com/snakers4/silero-models#text-to-speech - https://habr.com/ru/post/54948
34.
▲
High-Quality Text-to-Speech Running on 1 CPU Core Locally
(twitter.com)
2 points
by
snakers41
6y ago
|
1 comments
35.
▲
by
snakers41
6y ago
Free, high-quality speech-to-text running fast on one CPU thread. Not middleware, no cloud required, no GAFA APIs. 10 Voices in 5 languages. Can be run locally on your PC / device / phone.
36.
▲
by
snakers41
6y ago
this is hilarious we did not catch the error in the title (fixed)
37.
▲
High-Quality Text-to-Speech Made Accessible, Simple and Fast
(habr.com)
9 points
by
snakers41
6y ago
|
2 comments
38.
▲
by
snakers41
6y ago
Open (A-GPL), high quality (4 - 4.5 MOS) and fast (faster than real-time on 1 CPU thread) TTS models published: 10 voices, 5 languages, 8 kHz and 16 kHz models
39.
▲
by
snakers41
6y ago
also, importantly, not from corporations and not about GPT-3 xDD
40.
▲
by
snakers41
6y ago
Other languages are planned and are on the way, albeit w/o any hard deadlines - French, Italian, Polish, Czech
41.
▲
by
snakers41
6y ago
Soumith Chintala also briefly commented on the release - https://github.com/pytorch/hub/pull/153#pullrequestreview-49... , which is very inspiring \(ᵔᵕᵔ)/
42.
▲
Silero Speech-To-Text models
(pytorch.org)
36 points
by
snakers41
6y ago
|
9 comments
43.
▲
by
snakers41
6y ago
Benchmarks - https://github.com/snakers4/silero-models/wiki/Quality-Bench...
44.
▲
End-to-end Speech Recognition in PyTorch trained on 50kh of speech is released
(github.com)
4 points
by
snakers41
6y ago
|
0 comments
45.
▲
Russian hackers share their ASR models better than Google?
(github.com)
5 points
by
snakers41
6y ago
|
2 comments
46.
▲
by
snakers41
6y ago
Oh yeah, Russians with APGL license, but licenses do not work anyway
47.
▲
by
snakers41
7y ago
Hi, my name is Alexander, I one of authors of both Gradient pieces, Open STT and silero.ai. We are planning on adding 2-3 new languages soon (English, German, maybe Spanish), so if you would like us to fully open source everything - please
48.
▲
by
snakers41
7y ago
Hi, my name is Alexander, I am an author of both Gradient pieces, Open STT and silero.ai Interesting We did mostly the same Did you open source your data as well?
49.
▲
Building client routing / semantic search at Profi.ru
(spark-in.me)
1 points
by
snakers41
8y ago
|
1 comments
50.
▲
SpaceNet 3 Road Detector challenge solution
(spark-in.me)
1 points
by
snakers41
8y ago
|
0 comments
51.
▲
Parsing Common Crawl in 2 plain scripts in python
(spark-in.me)
2 points
by
snakers41
8y ago
|
0 comments
52.
▲
Parsing Wikipedia in 4 plain commands in Python
(spark-in.me)
4 points
by
snakers41
8y ago
|
0 comments