Talkie

54 posts

Talkie banner
Talkie

Talkie

@usetalkieapp

Voice + AI - Your Unified Voice System by @arach

@localhost Katılım Ocak 2026
9 Takip Edilen9 Takipçiler
Talkie
Talkie@usetalkieapp·
Talkie tweet media
ZXX
0
0
0
2
Talkie
Talkie@usetalkieapp·
✨ this could be on your phone ✨
Talkie tweet media
English
0
0
1
8
Talkie
Talkie@usetalkieapp·
`𝚗𝚙𝚡 @𝚝𝚊𝚕𝚔𝚒𝚎/𝚊𝚙𝚙 𝚒𝚗𝚜𝚝𝚊𝚕𝚕` ✨
English
0
0
1
13
Talkie retweetledi
Wildminder
Wildminder@wildmindai·
So many TTS models dropping lately, it's hard to keep up. Here you go. A curated list of open-source models for music, TTS, ASR, and Audio SR to save your sanity github.com/wildminder/awe…
Wildminder tweet media
English
4
40
329
16.3K
Talkie retweetledi
Prince Canuma
Prince Canuma@Prince_Canuma·
Voxtral-TTS by @MistralAI now faster than realtime on MLX-Audio streaming New release in a few!
Prince Canuma tweet media
English
10
11
164
9.3K
Talkie
Talkie@usetalkieapp·
Talkie now has a helpful terminal and it can help you build powerful workflows, no need to drag and drop ✨ Powered by OpenCode and Ghostty
Talkie tweet media
English
0
0
0
24
Talkie retweetledi
Cohere
Cohere@cohere·
Introducing: Cohere Transcribe – a new state-of-the-art in open source speech recognition.
English
82
297
2.6K
607.8K
Talkie retweetledi
Chubby♨️
Chubby♨️@kimmonismus·
Mistral AI released Voxtral TTS, a 3-billion-parameter text-to-speech model with open weights that the company says outperformed ElevenLabs Flash v2.5 in human preference tests roughly 63% of the time on standard voices and nearly 70% on voice customization. The model runs on about 3 GB of RAM, achieves 90-millisecond time-to-first-audio, supports nine languages, and can clone a voice from just five seconds of reference audio, including cross-lingual adaptation that preserves the speaker's accent.
Chubby♨️ tweet mediaChubby♨️ tweet media
English
44
115
1.3K
80.5K
Talkie
Talkie@usetalkieapp·
There are rumors that Talkie is adding a beautiful terminal with the best dictation models :)
Talkie tweet media
English
0
0
1
9
Talkie
Talkie@usetalkieapp·
@SeanOliver @claudeai @WisprFlow There's a real benefit to using the same input across apps. If anything, Claude Code voice mode and all the different voice inputs are great when you're on a device without an app installed. once you're voice pilled, you need tools that are optimized for advanced voice workflows.
English
1
0
1
22
Sean Oliver
Sean Oliver@SeanOliver·
Been testing @claudeai's new voice feature and @WisprFlow side by side. Right now? Wispr Flow wins. Claude voice is great for what it does. But it only does one thing: prompt Claude. That's pretty niche. Wispr gives you voice control across every app. Dictation everywhere. That's the actual unlock. The thing about voice interfaces is once you start using them, you want them for everything. Not just AI prompts. Emails. Docs. Linear tickets. Slack messages. A standalone app that only talks to Claude feels like solving 5% of the problem. Claude voice will get better for sure. But I'm curious how many people will actually use it long-term vs. just going full system-wide voice. The ROI on learning a new voice workflow only makes sense if it works everywhere.
English
1
0
1
87
Talkie
Talkie@usetalkieapp·
Continuing to teach Talkie how to hear bash. Now training on a mac mini. Could also be titled: "what happens when generalists vibe tune models" usetalkie.com/ideas/training…
English
0
0
0
26
Talkie
Talkie@usetalkieapp·
@yenkel @zachtratar I super agree with you, and I think this new flow *can* converge to better ideas faster (for people who care)
English
0
0
2
19
yenkel
yenkel@yenkel·
main point is BDUF doesn't help (imo never did) and AI makes the cost of protos so low that there is no longer an excuse you can prototype for a while and then you can right a design that mixes eng and product concepts (a lot of PRDs stayed vague) that is useful for people and AI and of course keep it up to date with AI so do you at some point write a document with requirements? yes is it how you typically did a PRD and what you included? imo, no. and you are recommended to arrive at it differently do you still need to talk to customers? yes ofc. and show them the prototypes. asap. how to do this at scale is the new question imo x.com/yenkel/status/…
English
2
0
8
2K
Talkie
Talkie@usetalkieapp·
@yenkel The new PRD is you talk for a few minutes about what you want in Talkie and run a few prompts on your memos until they are ready for Claude code it's like `/voice` but with better UX :)
English
1
0
1
46
yenkel
yenkel@yenkel·
you must internalize this ASAP: - less handoffs, decide fast - faster exploration - encourage to throw away code/tokens - learn by building, de risk with code - pick leads that can own design, eng and product
yenkel tweet media
Gergely Orosz@GergelyOrosz

What does it mean for software engineering when we no longer write the code? Here's the take from Boris Cherny (@bcherny), the creator of Claude Code. Timestamps: 00:00 Intro 11:15 Lessons from Meta 19:46 Joining Anthropic 23:08 The origins of Claude Code 32:55 Boris's Claude Code workflow 36:27 Parallel agents 40:25 Code reviews 47:18 Claude Code's architecture 52:38 Permissions and sandboxing 55:05 Engineering culture at Anthropic 1:05:15 Claude Cowork 1:12:48 Observability and privacy 1:14:45 Agent swarms 1:21:16 LLMs and the printing press analogy 1:30:16 Standout engineer archetypes 1:32:12 What skills still matter for engineers 1:35:24 Book recommendations Brought to you by: • @statsig  — ⁠ The unified platform for flags, analytics, experiments, and more. statsig.com/pragmatic • @SonarSource – The makers of SonarQube, the industry standard for automated code review. Proactively find and fix issues in real-time with the SonarQube MCP Server: sonarsource.com/products/sonar… • @WorkOS – Everything you need to make your app enterprise ready. workos.com Three interesting things from this conversation: 1. Boris automated himself out of code review well before AI. Boris was one of the most prolific code reviewers at Meta company. And he worked hard to minimize time spent on code review. His system::every time he left the same kind of review comment, he logged it in a spreadsheet. Once a pattern hit 3-4 occurrences, he’d write a lint rule to automate it away! 2. PRDs are dead on the Claude Code team: prototypes replaced them. Instead of writing Product Requirement Documents (specs), they build hundreds of working prototypes before shipping a feature. Boris: “There’s just no way we could have shipped this if we started with static mocks and Figma or if we started with a PRD.” 3. This is the year of the generalist (and maybe the year of those with ADHD) Boris’s work has shifted from deep-focus single-threaded coding to managing multiple parallel agents and context-switching rapidly. As Boris put it: “It’s not so much about deep work, it’s about how good I am at context switching and jumping across multiple different contexts very quickly.”

English
49
97
1.5K
423.4K
Talkie
Talkie@usetalkieapp·
Hi 
GIF
1
0
1
15
Talkie
Talkie@usetalkieapp·
@frantzfries API and CLI with a clean authentication model
English
0
0
0
6
Chris Frantz
Chris Frantz@frantzfries·
everything in the GUI should be available in the API this will be the theme for us for the remainder of the year the UI becomes the preview layer for agentic work if you can click it, you should be able to prompt it
English
33
18
203
13.8K
Talkie
Talkie@usetalkieapp·
@yongfook give Talkie a spin, comes with an agent friendly CLI `𝚌𝚞𝚛𝚕 -𝚏𝚜𝚂𝙻 𝚐𝚘.𝚞𝚜𝚎𝚝𝚊𝚕𝚔𝚒𝚎.𝚌𝚘𝚖/𝚒𝚗𝚜𝚝𝚊𝚕𝚕 | 𝚋𝚊𝚜𝚑` private, local and incredibly fast
English
1
0
0
187
Jon Yongfook
Jon Yongfook@yongfook·
I used to think the idea of voice-dictating to Claude or other AI tools was dumb, but the more comfortable I get the more I want to do it. Obviously I wouldn't do it in a public setting, but at home, cup of coffee in hand, I'm like... hmm I should just talk instead of type.
English
68
3
154
34.6K