tweetindex
DE

Inworld AI ✓

@inworld · Mountain View, CA · joined 17 Aug 2021

The Realtime AI Company. Powering the next generation of consumer-facing applications.

12 641Followers
204Following
543Posts total
1.2MViews on collected posts

Neueste Beiträge

@inworld @altsoph https://t.co/R68DVqUQwJ 6 views · 0 likes · 0 reposts · 0 replies Open on X →
@inworld @altsoph It’s great that they share tools like this 28 views · 0 likes · 0 reposts · 0 replies Open on X →
@inworld @altsoph Open-sourcing evaluation tools like this is a great step toward more transparent and reliable TTS benchmarking 20 views · 1 likes · 0 reposts · 0 replies Open on X →
@inworld Beautiful, just beautiful https://t.co/ASDEwGCwT6
615 views · 10 likes · 0 reposts · 0 replies Open on X →
Realtime TTS-2 allows natural language voice direction to instruct how every line should be spoken. You decide the delivery. [speak through gritted teeth, barely holding it in] “I'm happy for you. Really.” [speak warmly, like you mean every word] “I'm happy for you. Really.” 2.1K views · 22 likes · 0 reposts · 1 replies Open on X →
With Realtime TTS-2, you feel like a director working with a supernatural voice actor - allowing new levels of control and responsiveness. @livekit has been a close partner on TTS-2, with CTO David Zhao calling it “a real step forward in emotionally expressive voice synthesis.” 2K views · 19 likes · 0 reposts · 1 replies Open on X →
Realtime TTS-2 Features: → Natural language voice direction → ~100ms time-to-first-byte → Crosslingual support: same identity in 200+ languages → Self-serve professional voice cloning → Natural language voice design → Multi-turn context → $12.50/Mchar (1.5c per min) on sub 2.5K views · 25 likes · 0 reposts · 1 replies Open on X →
Hundreds of millions of people already interact with Realtime TTS-2 every day through apps built on the model. It powers most of the top 1% of consumer apps across language learning, tutoring, companions, roleplay, wellness, fitness, news, games and CX. 3K views · 24 likes · 0 reposts · 1 replies Open on X →
Realtime TTS-2 Flash is also launching today as the fastest text-to-speech in the world with 25ms time-to-first-byte. Half the price and topping leaderboards, ahead of models 20x the cost and latency. Built for applications where every millisecond matters. 5.7K views · 46 likes · 0 reposts · 1 replies Open on X →
Realtime TTS-2 is GA today, now the #1 model on Artificial Analysis and fastest TTS in the world of its class. The research preview got far more usage than expected. That experience went back into research and today's GA model represents the biggest leap in Inworld's history. ht
2:28
1.2M views · 1.1K likes · 235 reposts · 172 replies Open on X →
Our Head of ML/AI Research, @altsoph just open-sourced our TTS evaluation toolkit. TTS evaluation is less standardized than people assume. Two teams can run the same model on the same dataset and report different word error rates. The ASR model you measure with moves the number. 4K views · 68 likes · 7 reposts · 15 replies Open on X →
Singapore AI builders, this one's for you. Inworld is teaming up with the SMU Artificial Intelligence Club and Pathwise Academy for an AI Voice Builder Night at Singapore Management University. This won't be an evening of sitting through slides. We're getting hands-on. What 1.8K views · 15 likes · 0 reposts · 6 replies Open on X →
Last week we co-hosted the Voice Agent Build Day at @agihouse_org. One day, ~100 builders, one rule: ship something you can talk to. 27 projects made it to demos. Among them: - A poker game where AI billionaires trash talk you between hands - A Mandarin tutor that remembers http
2:02
2.1K views · 16 likes · 2 reposts · 4 replies Open on X →
Our Head of Research, @altsoph, PhD wrote up something that we thought was pretty cool. He wanted to stress-test our TTS models with tongue twisters, but the classic ones don't really work, because the models were trained on them. So he got an agent to make up new ones by 1.4K views · 10 likes · 2 reposts · 0 replies Open on X →

Im Vergleich zu Konten gleicher Größe

14 Beiträge aus den letzten 90 Tagen, verglichen mit der Größenklasse 10K–100K Follower. wird weit gezeigt, aber nur wenige dieser Zuschauer reagieren.

Medianaufrufe2 030dieses Konto924Median für 10K–100K
Reichweite, %16.06%dieses Konto3.62%Median für 10K–100K
Interaktion, %1.03%dieses Konto1.52%Median für 10K–100K
KennzahlDieses KontoMedian für 10K–100KVerhältnis
Medianaufrufe pro Beitrag2 0309242.20×
Reichweite (Aufrufe ÷ Follower)16.06%3.62%4.44×
Interaktionsrate1.03%1.52%0.68×

Weitere Konten dieser Größe →   Mit einem anderen Konto vergleichen →   Wie diese Vergleichswerte entstehen →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

1.4K6 Aug
2.1K10 Aug
1.8K28 Aug
4K31 Aug
1.2M2 Sep
5.7K
3K
2.5K
2K
2.1K
615
203 Sep
28
66 Sep

Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.88%6 Aug
1.06%10 Aug
1.19%28 Aug
2.25%31 Aug
0.14%2 Sep
0.84%
0.82%
1.05%
1.00%
1.10%
1.63%
5.00%3 Sep
0.00%
0.00%6 Sep

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.

What the audience does

Likes45.6%1 385 in total
Reposts8.1%246 in total
Replies6.6%202 in total
Quotes5.7%174 in total
Bookmarks33.9%1 031 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

Ähnliche Konten