Active group: 27 people wrote in 30 days, 79% of them replied to others, 30% of messages get a reply, audience is stable.
About the group
Discussion of speech recognition, TTS, audio separation libraries, and ML models.
Description in Telegram
Group about speech technology news
Key numbers
1 774
members
24 h±0
7 days±0
30 days−31
268
messages in 30 days
≈ 8.9 a day
indexed through Nov 5
27
writing in 30 days
Catalog rank: #1 679 971 · #112 389 in language
Audience
Members 1,774
history since 2026-10-03
2026-10-032026-11-07
07 days
−3130 days
Is this your group? Show its statisticsA line for the description or an ad post and a badge for your website — with fresh numbers and a link to this page.
Get the link and badge
Put the line in the description and the “✓ linked by owner” mark will appear next to the name on the site, in the bot and in the mini app. We re-read descriptions ourselves; to check right away, use “Check the link” in the bot or mini app.
For the Telegram description
HTML for a website
Markdown — README and wikis
BBCode — forums
Posts
Latest posts
Looks like it is a great test for liveness to ask voice agent to sing a bitFor ASR models under 150 MB size, I feel the Parakeet-derived Phonon-2 ASR model is impressive for on-device inference, achieving a WER of around 5.34% despite its compact size. Are there any smaller on-device ASR models that can outperform…Thanks and looks like feasible solution to keep it on mobile devicesCan anyone suggest best on device diarization model that can run on Apple/android phone?the issue is finding out whether it can beat an initialized decoder or not. especially at scalei'm treating it as some sort of proxy to assess its language understandingAll posts →Similar venues — ML & research
M
Machine Learning with Python
Computer Science and Programming
Machine Learning
Aethir Announcements
Python Projects & Resources
AI with Papers - Artificial Intelligence & Deep Learning
@CodeProgrammer
channel68 474 subscribers11 posts/wk1 authors/wkenC@computer_science_and_programming
channel139 419 subscribers1 posts/wk1 authors/wkenM@MachineLearning9
channel41 899 subscribers12 posts/wk1 authors/wkenA@Aethir_Official
channel127 195 subscribers3 posts/wk1 authors/wkenP@pythondevelopersindia
channel63 480 subscribers6 posts/wk1 authors/wkenA@AI_DeepLearning
channel17 260 subscribersenHistory
Name history
what this venue was called before and which @usernames it held — from the index
06.11.2026first seen · «Speech Recognition» @speech_recognition
27.03.2019upgraded from the basic group «Speech Recognition» · from the venue's own feed
Archive by month · 93
Archive by month
November 2026October 2026September 2026August 2026July 2026June 2026May 2026April 2026March 2026February 2026January 2026December 2025November 2025October 2025September 2025August 2025July 2025June 2025May 2025April 2025March 2025February 2025January 2025December 2024November 2024October 2024September 2024August 2024July 2024June 2024May 2024April 2024March 2024February 2024January 2024December 2023November 2023October 2023September 2023August 2023July 2023June 2023May 2023April 2023March 2023February 2023January 2023December 2022November 2022October 2022September 2022August 2022July 2022June 2022May 2022April 2022March 2022February 2022January 2022December 2021November 2021October 2021September 2021August 2021July 2021June 2021May 2021April 2021March 2021February 2021January 2021December 2020November 2020October 2020September 2020August 2020July 2020June 2020May 2020April 2020March 2020February 2020January 2020December 2019November 2019October 2019September 2019August 2019July 2019June 2019May 2019April 2019March 2019
catalog data as of 2026-11-06 04:25 UTC · How we count →