Congrats Recall team, I've been a customer for the past 1.5 years and the entire infra layer behind meetings has let our company really focus on "what makes our beer taste better" instead of having to worry about building universal support for different (and tedious) platforms like Teams and Zoom. Eager to give the desktop recording sdk a try soon.
Our meetings often involve a mix of onsite and offsite employees. Typical setup might be CEO + CTO + a VP in a room, connected as a single zoom client to the call (either of these 3 guys depending on who got in the meeting room first), then few additional people joining remotely from home each on their own zoom instance. The guys on the meeting room are using a dedicated camera in the meeting room that captures the entire room, and has all participants in sight.
Is this a setup you are trying to address; how are you able to recognize speakers in this configuration ?
Most transcript system we have tried bundle everything that is said by the onsite people as a single entity which pretty much destroys the value of the transcript; especially if people in that room disagree with each other; reading the transcript makes it feel that the onsite guys is very schizophrenic
Have you explored using speaker diarization and speaker identification, given that pyannote etc. takes this approach?
I'm curious given your decision to capture speaker names from the screen. I see the merits during desktop recording, but I can also see how this limits utility when trying to offer the same functionality across desktop and other scenarios (e.g. in-person meetings, audio uploads etc.)
We already support diarization in the Desktop Recording SDK by capturing the meeting platform’s speaker-change events, so you get a diarized transcript plus precise “speaker started talking” timestamps out of the box. We also support voice-signature diarization via third-party STT providers for participants calling in from the same room
For in-person meetings and audio uploads, this is on our roadmap and in development. More to come on this!
> we support 0-day retention, and we don’t train models on customer data.
Checks out the website[1]:
> By default, all media associated with a recording is retained indefinitely. If needed, you can request early deletion of this data at any time via our API.
Data shared with 25 "subprocessors", some of who also retain data indefinitely. Yikes!
I'm impressed with the desktop SDK demo video hosted by Nick. Very clever. I noticed he's using Emacs, and that got me thinking that maybe I could make a little capture template that invokes your service to transcript directly from org mode. :) - Adding this to the wood pile.
You're right, and I agree that participants should be aware when they’re being recorded
Because consent laws are complex and vary by region and industry, we leave the consent flow to the developer and we provide the tools and guidance to do it correctly. As with our Meeting Bot API, we also urge teams to follow local laws and make recording clearly visible to users
Wow, congrats on finally using up your single Launch HN, David and Amanda! :wink:
No but seriously, y'all have built not only an incredible product that I had the chance to demo, but a great company as well, through your previous pivots and cofounder changes. You're building schlep tools that product companies _definitely_ don't want to do, years before it was clear there was a market here, and do it well.
There's definitely demand for a native screen recorder, and I think it's the right move to be agnostic to privacy (the lower down the stack you go, the more permissable you should be about use-cases). Imagine how much competition in file storage there would have been had there been an API provider for Dropbox's Finder sync technology (though you could argue it just incentivizes large companies like Hubspot to build their own screen recording feature into their platform, rather than enabling new startups like Gong but I digress).
Y'all deserve the success that you have, and wishing you all the best of luck with the new product launch!
Congrats on ur launch. Amanda has the strongest LinkedIn game I have ever seen in my life. On the other hand the product is IMHO at risk?
Models like Whisper, DistilWhisper, TinyLlama, miniGPT-4, OpenHermes, Vosk, and Llama.cpp make Recall.ai meeting transcription easy to replicate. IMHO in 1 weekend you can build an open-source tech stacks that can rival or EVEN surpass the value brought....or am I tripping?
Customer here, you're tripping. Recall provides transcription as an auxiliary service, not their core value prop.
Recall is, at its core, an API for bot recording. As someone building an application that relies heavily on conversational data, recording meetings is really important. Recall makes that process as easy as an API call, standardized across various meeting platforms. It's a huge PITA to set up infrastructure to get bots to join meetings that handle each platforms' proclivities, encoding and storing video data, etc.
The transcription service is just something they do to make transcribing recordings - one of the most common first post-processing steps for any conversational data - easier and lower friction.
I actually agree that it’s become incredibly easy to transcribe conversations using open-source models, and that’s not where Recall adds the most value. The hard part is building the infrastructure that allows you to get real-time access to the raw audio, video, and transcript data directly from the meeting platforms. We abstract all of that away and provide you with a clean interface to access that data. Once you get the data, you could use any of the models that you mentioned to do your own transcription, or transcribe using Recall’s transcription models.
Ah yes, the ever popular "over a weekend" retort. So, you have a weekend coming up. I fully expect to see your Show HN on Monday. You'd know pretty quickly on your own if you were tripping or not. I'll check back with you on Monday. I'll gladly eat a bowl of something (I'm thinking ice cream) if you have a working Show HN on Monday. I'm giving you two days of prep. Or you can take the same weekend time and provide a Show HN on Saturday. Choice is yours
I’m interested in how you expect to keep market share if this capability can be offered by the webrtc service provider. I saw the other comment about multiple providers, but many enterprises have just one preferred path. For example I’ve been recording Google drive calls and the transcript goes straight into my Google drive.
For internal use cases like recording your own meetings into Google Drive, the native tools work fine.
Where we come in is for companies building products that need to support all of their customers across Zoom, Meet, Teams, Webex, etc. Most enterprises don’t want five different integrations, and native APIs often come with restrictions (like only the organizer being able to access the file, or recordings not being available until after the call).
Congrats on the launch! I'm working on a new tool for startups sales (https://closer.so) and in many customer interviews the point of not wanting the bot in the meeting kept coming up. I love how Recall keeps brining frontier tech as APIs
I have Loom recorded a Zoom meeting so get it. I think for corporates though the integrated approaches are so so convenient. Have your meeting and get a summary email by doing nothing (or one click opt in). I feel like your solution is for edge cases where the mainstream ways are not possible.
Just to clarify, we’re the infra layer that reliably captures and normalizes meeting data across platforms. The real value for users is what developers build on top: automated analysis, enrichment, and workflows (not the capture itself)
Modern LLMs can power sales coaching, medical scribing, legal review, support QA, and compliance reporting but they need consistent inputs to process. We handle capture/formatting/edge cases so developers can focus on models and UX
Related but tangential: is there a way to grab video frames from a meeting? Audio transcript is great. Looking at a use case to grab participant video.
> Back in W20, our first product was an API that lets you send a bot participant into a meeting. This gives developers access to audio/video streams and other data in the meeting. Today, this API powers most of the meeting recording products on the market.
The median salary in the US is $29/hour.
By definition a one hour meeting has at least two people in it; often more. So two median guys talking for an hour costs ~$60. The meeting the you really want transcripts for often contain more than one person; and often involve people earning more than the median. I'd happily ad $1 to every single one of my meetings if they get more productive.
It is a lot but processing real time video and audio streams inherently consumes alot of CPU. So they may not be making as much profit on that price as you'd think.
I run an open source alternative to Recall (for meeting bots), and our costs are about 8 cents per hour.
$0.70/hr is our starter rate for low-volume testing. In production, developers will see higher usage and choose to commit to volume and longer-term usage. Because of this, we've seen most teams don’t pay the starter price once they scale beyond early pilots
Enabling transcription/recordings per platform and remembering to record creates user-dependent setup. Also the host often needs to install apps which adds security friction, and you still have to build/maintain separate implementations for Zoom/Meet/Teams which is often a cost that devs don't want to deal with
Instead, we built a single API that can get the same results without the issues mentioned above so you can focus on building the features your users care about
Congrats Recall team, I've been a customer for the past 1.5 years and the entire infra layer behind meetings has let our company really focus on "what makes our beer taste better" instead of having to worry about building universal support for different (and tedious) platforms like Teams and Zoom. Eager to give the desktop recording sdk a try soon.
Thanks and love to hear this!
Our meetings often involve a mix of onsite and offsite employees. Typical setup might be CEO + CTO + a VP in a room, connected as a single zoom client to the call (either of these 3 guys depending on who got in the meeting room first), then few additional people joining remotely from home each on their own zoom instance. The guys on the meeting room are using a dedicated camera in the meeting room that captures the entire room, and has all participants in sight. Is this a setup you are trying to address; how are you able to recognize speakers in this configuration ?
Most transcript system we have tried bundle everything that is said by the onsite people as a single entity which pretty much destroys the value of the transcript; especially if people in that room disagree with each other; reading the transcript makes it feel that the onsite guys is very schizophrenic
Out of interest, what is the thinking behind sending a physical mailer to what feels like a large fraction of San Francisco?
Both why send it and why send it with very little info included on the page?
Did you get one? :) This was a part of our Series B raise to help get our name out
I did. Were you trying to get VCs to know about you? Or more like a twilio billboard that just said ask your devs?
The postcards were part of a dev-focused campaign to get people curious enough to check us out. We kept it minimal to stand out amongst other mail.
Wait, they sent out physical spam?
Have you explored using speaker diarization and speaker identification, given that pyannote etc. takes this approach?
I'm curious given your decision to capture speaker names from the screen. I see the merits during desktop recording, but I can also see how this limits utility when trying to offer the same functionality across desktop and other scenarios (e.g. in-person meetings, audio uploads etc.)
We already support diarization in the Desktop Recording SDK by capturing the meeting platform’s speaker-change events, so you get a diarized transcript plus precise “speaker started talking” timestamps out of the box. We also support voice-signature diarization via third-party STT providers for participants calling in from the same room
For in-person meetings and audio uploads, this is on our roadmap and in development. More to come on this!
> we support 0-day retention, and we don’t train models on customer data.
Checks out the website[1]:
> By default, all media associated with a recording is retained indefinitely. If needed, you can request early deletion of this data at any time via our API.
Data shared with 25 "subprocessors", some of who also retain data indefinitely. Yikes!
[1]: https://security.recall.ai/
There's no contradiction. One statement is about the default, the other is about the possibility.
The contradiction is that some of those sub-processors retain data indefinitely[1], with no possibility of deletion.
[1]: https://arstechnica.com/tech-policy/2025/06/openai-confronts...
I'm impressed with the desktop SDK demo video hosted by Nick. Very clever. I noticed he's using Emacs, and that got me thinking that maybe I could make a little capture template that invokes your service to transcript directly from org mode. :) - Adding this to the wood pile.
Pardon my ignorance, but is recording a call without informing the other participants considered bad practice?
Congrats on the launch! :tada:
You're right, and I agree that participants should be aware when they’re being recorded
Because consent laws are complex and vary by region and industry, we leave the consent flow to the developer and we provide the tools and guidance to do it correctly. As with our Meeting Bot API, we also urge teams to follow local laws and make recording clearly visible to users
Thanks for the clarification.
It's not just bad practice, it's illegal in many countries. It is in France for instance - but it's not like you can actively prevent it
It's illegal in some states, legal in others.
Consult Linda Tripp
Wish she was around!
Also wondering this.
Wow, congrats on finally using up your single Launch HN, David and Amanda! :wink:
No but seriously, y'all have built not only an incredible product that I had the chance to demo, but a great company as well, through your previous pivots and cofounder changes. You're building schlep tools that product companies _definitely_ don't want to do, years before it was clear there was a market here, and do it well.
There's definitely demand for a native screen recorder, and I think it's the right move to be agnostic to privacy (the lower down the stack you go, the more permissable you should be about use-cases). Imagine how much competition in file storage there would have been had there been an API provider for Dropbox's Finder sync technology (though you could argue it just incentivizes large companies like Hubspot to build their own screen recording feature into their platform, rather than enabling new startups like Gong but I digress).
Y'all deserve the success that you have, and wishing you all the best of luck with the new product launch!
Thanks! Really appreciate the kind words
Congrats on ur launch. Amanda has the strongest LinkedIn game I have ever seen in my life. On the other hand the product is IMHO at risk? Models like Whisper, DistilWhisper, TinyLlama, miniGPT-4, OpenHermes, Vosk, and Llama.cpp make Recall.ai meeting transcription easy to replicate. IMHO in 1 weekend you can build an open-source tech stacks that can rival or EVEN surpass the value brought....or am I tripping?
Customer here, you're tripping. Recall provides transcription as an auxiliary service, not their core value prop.
Recall is, at its core, an API for bot recording. As someone building an application that relies heavily on conversational data, recording meetings is really important. Recall makes that process as easy as an API call, standardized across various meeting platforms. It's a huge PITA to set up infrastructure to get bots to join meetings that handle each platforms' proclivities, encoding and storing video data, etc.
The transcription service is just something they do to make transcribing recordings - one of the most common first post-processing steps for any conversational data - easier and lower friction.
Amanda says thank you so much!
I actually agree that it’s become incredibly easy to transcribe conversations using open-source models, and that’s not where Recall adds the most value. The hard part is building the infrastructure that allows you to get real-time access to the raw audio, video, and transcript data directly from the meeting platforms. We abstract all of that away and provide you with a clean interface to access that data. Once you get the data, you could use any of the models that you mentioned to do your own transcription, or transcribe using Recall’s transcription models.
Ah yes, the ever popular "over a weekend" retort. So, you have a weekend coming up. I fully expect to see your Show HN on Monday. You'd know pretty quickly on your own if you were tripping or not. I'll check back with you on Monday. I'll gladly eat a bowl of something (I'm thinking ice cream) if you have a working Show HN on Monday. I'm giving you two days of prep. Or you can take the same weekend time and provide a Show HN on Saturday. Choice is yours
I’m interested in how you expect to keep market share if this capability can be offered by the webrtc service provider. I saw the other comment about multiple providers, but many enterprises have just one preferred path. For example I’ve been recording Google drive calls and the transcript goes straight into my Google drive.
For internal use cases like recording your own meetings into Google Drive, the native tools work fine.
Where we come in is for companies building products that need to support all of their customers across Zoom, Meet, Teams, Webex, etc. Most enterprises don’t want five different integrations, and native APIs often come with restrictions (like only the organizer being able to access the file, or recordings not being available until after the call).
Congrats on the launch! I'm working on a new tool for startups sales (https://closer.so) and in many customer interviews the point of not wanting the bot in the meeting kept coming up. I love how Recall keeps brining frontier tech as APIs
Hello are you open to hiring for Remote roles ?? am looking for Full stack junior or intern roles and was truely impressed by the product
I have Loom recorded a Zoom meeting so get it. I think for corporates though the integrated approaches are so so convenient. Have your meeting and get a summary email by doing nothing (or one click opt in). I feel like your solution is for edge cases where the mainstream ways are not possible.
Just to clarify, we’re the infra layer that reliably captures and normalizes meeting data across platforms. The real value for users is what developers build on top: automated analysis, enrichment, and workflows (not the capture itself)
Modern LLMs can power sales coaching, medical scribing, legal review, support QA, and compliance reporting but they need consistent inputs to process. We handle capture/formatting/edge cases so developers can focus on models and UX
Hi David and Amanda! Followed you guys from the very beginning, glad to see Recall.ai get so big!
Related but tangential: is there a way to grab video frames from a meeting? Audio transcript is great. Looking at a use case to grab participant video.
> Back in W20, our first product was an API that lets you send a bot participant into a meeting. This gives developers access to audio/video streams and other data in the meeting. Today, this API powers most of the meeting recording products on the market.
> Desktop Recording SDK
So this wouldn't work in a web app, only a desktop app?
I guess that leaves electron and tauri?
Is it a good idea to stay named like the spyware that MS has put in Windows 11?
98% of people don’t know. Including me and I am here often.
Granola?
70 cents per hour is a mountain of fees... basically a $1 per meeting. Sheesh.
The median salary in the US is $29/hour. By definition a one hour meeting has at least two people in it; often more. So two median guys talking for an hour costs ~$60. The meeting the you really want transcripts for often contain more than one person; and often involve people earning more than the median. I'd happily ad $1 to every single one of my meetings if they get more productive.
It is a lot but processing real time video and audio streams inherently consumes alot of CPU. So they may not be making as much profit on that price as you'd think.
I run an open source alternative to Recall (for meeting bots), and our costs are about 8 cents per hour.
What is the open source project?
https://attendee.dev/
$0.70/hr is our starter rate for low-volume testing. In production, developers will see higher usage and choose to commit to volume and longer-term usage. Because of this, we've seen most teams don’t pay the starter price once they scale beyond early pilots
Is this with active speech or you pay in every second of silence too?
Usage includes silent time too as we are still processing the media streams
What is the value in paying this massive cost when Teams, Zoom all support ootb?
Enabling transcription/recordings per platform and remembering to record creates user-dependent setup. Also the host often needs to install apps which adds security friction, and you still have to build/maintain separate implementations for Zoom/Meet/Teams which is often a cost that devs don't want to deal with
Instead, we built a single API that can get the same results without the issues mentioned above so you can focus on building the features your users care about
[dead]