ElevenLabs CEO: Why Voice is the Next AI Interface

We don't want to become same as previous

generation of of the editing suite. So

instead, let's solve it on the research

level where it will know based on the

voice exactly how it should speak with

the speed. To be able to cut to all

those different use cases, you need such

a big array of different voices,

different languages, different accents,

um different styles. Um so we launched

voice marketplace where you you could

create your voice and then uh share it.

And when that voice is shared, you earn

money in the return. Today we have

almost 10,000 voices. We paid $10

million back to the people in the

community. There's some crazy stories

from the voices. Just speaking through

exactly the technology, showing the

examples and kind of avoiding this

initial knee-jerk reaction that AI is

bad has has been has been tremendous.

[Music]

>> Um, I'm excited to welcome our first

speaker, Maddie, co-founder and CEO of

11 Labs.

[Music]

[Applause]

All right. So, good to have you here,

Maddie.

>> Thanks so much for having me here. Great

great to see everyone and good morning.

>> And that

>> that was uh the walkcom music generated

by 11 Labs. Was it?

>> It was. We we we expand continuously

across the audio space. So we started

with voices then created orchestration

of how to build voice agents and now

also create a fully licensed music model

so can produce amazing music to go

alongside with it.

>> Awesome. We'll talk about all all about

that. I've um had the opportunity and

also uh the luck to get to know from the

very early days when 11 Labs got started

and get to partner over the last three

years to just see your execution

everywhere from product launches to

shipping new lines and models like you

just mentioned everything from uh text

to speech models speech to text and then

we uh started uh doing music sound

effects and now the AI agent platform um

I'm very curious first I'm still in awe

of the the shipping speed uh after all

the three years. Um but I want to ask

how do you actually maintain both the

speed and quality when you have such

expensive product road map?

>> So f first of all we we partnered almost

three years ago and um so it's it's uh

great to hear all the kind notes but

also what they didn't realize when we

partnered the infrastructure team was

free people and of course now I'm 11

loves founder. love number 11 and the

company infra team is 11 people so we

seeing the growth of the other side as

well um and and I hear that the

companies here raised $66 billion in the

total fundraising so the number 11 is

everywhere here

but the I think that to start off I

think first piece I have a I think the

smartest person I I I got to know as my

co-founder P who who has been the

research brain for creating a lot of the

the models and then being able to

assemble what who we think are the most

incredible researchers in the voice

space to really create the first text to

speech model that could understand the

context in a better way and turn that

into the emotion inonation. Then find a

way to um capture the characteristics of

the voice. So you have the voice sound

uh uh with the right style with the

right age with the right gender dialect

everything in in one. Um and then the

researchers across of course now

expanded that to speech to text music

and other work. So that's our foundation

and then the way we structure it to be

able to ship quickly especially with so

many things uh happening in AI space is

a lot of small teams. So today we have

roughly 20 product teams each of five to

10 people size which with full

independence can go ahead and ship

products. Of course that carries some of

the um uh um sometimes issues of

duplicative work or sometimes uh people

going uh um at different speeds. But at

the positive end, the ownership of each

of the teams is extremely high. So

people know that this is down to them to

really deliver and ship. Um and it

allows us to move extremely quickly. Um

we bucket our work into creative space.

So creative platform where we help with

narrations, voiceovers, dubs for for

creatives and creatives in the in the

media entertainment space. Um and um and

then on the agent side where we help

people recreate voice agent experience,

conversational agent experience across

customer experience all the way through

to immersive immersive media.

>> Great. 11 Labs has a labs name in the

labs in the name uh very similar to many

of the other big labs which means you're

doing your first party R&D and model

development but also building all these

20 products. How do you think about

balancing both like keep progressing on

the model research but at the same time

not delaying sort of the product

launches? Yeah, it's very tricky. I'm

sure many of you have the same thing

like do you do you build a product while

uh when you don't know if the research

innovation will displace the product you

just built. We had this in the early

days too. So one of the simple examples

was we um we we had a model at work and

one of the most common requests was

could we do a different speeds for

voices. So could you have additional

slider to modify the speed of how audio

gets generated and how quickly it speaks

and we are very against this idea of

like no we don't want to do any sliders

any toggles we don't want to become same

as previous generation of of the editing

suite. So instead let's solve it on the

research level where it will know based

on the voice exactly how it should speak

with the speed and um and we resisted

this for I think good amount of nine

months and we couldn't solve it on the

research side and then the product was

super simple solve that got all the all

the users across and now the approach we

take and like looking at this is if we

think the research work will take more

than three months then the product is um

can do any any anything they want to to

start um adding other models adding some

of the extensions. Of course, sometimes

the timeline is is tricky to predict,

but roughly the guidance we have from

our internal research team. What are the

initiatives we hope to ship this

quarter? What are long-term initiatives?

And then for anything long-term, you can

use any any other work to close that gap

and make it and make it better.

I guess first uh you kind of have to

figure out if the research commitment is

going to meet the the timeline first and

then go on to align with the the product

teams. Um that that make a lot of sense.

Um, as everyone is moving to San

Francisco and building in person and

locked in, uh, like in the same space,

11 has always been building globally and

having people more distributed, but you

now have centers, I guess, in different

locations from London, Warsaw, San

Francisco to New York and and other

places. um how do you think about

building this global expansion and

finding talent um globally versus I

guess the trade-offs of building at in

the same place?

>> Yeah. And we so we started uh so me and

my co-ounder Polish we started between

Warso and London at the time. Um and we

we we I think 11 Labs wouldn't have

existed if we weren't starting from

Europe. Uh it's a very peculiar thing

but in Poland if you watch a movie in

Polish language like a foreign movie in

Polish language all the voices whether

that's a male voice or female voice get

narrated with one single character no

emotions no inonation as you can imagine

it's pretty terrible and it's still

happening today for most of the content

out there and

>> I've had a similar experience growing up

in China that we have a lot of western

movies dubbed in Chinese monotoned

>> so bad so bad and it's like in in Poland

of course postcommunist country, it's a

cheaper way to do it. You don't have to

hire as many people. You have one

monoton um audiobook reading of of a

movie. And um and that was kind of where

the company started. And we started

initially in in Europe. And we realized

that if we want the best people to solve

what was a research problem at the time,

we need to hire wherever they are. And

um we couldn't lock ourselves to just

San Francisco or uh or or look at the

West Coast. we we knew that we need to

find them across Europe, across Asia and

bring them into the company. So we

started fully remote and um and started

looking at those those those people and

then on engineering we also were very

against this uh traditional hiring

method of looking at LinkedIn looking at

traditional uh traditional background

and trying to figure out could we go and

and and figure out a different method to

to hire people that led to some very

interesting hires. So we hired a person

that uh had a incredible open-source

texttospech model and was working in the

call center at the same time as a

recipient of the calls to make money.

>> Wow. and uh and he's now in the team,

one of the most brilliant researchers we

have uh doing all the data processing.

But um the the the same pattern kind of

followed and of course the early team

was very distributed and then as we

started scaling so beyond 30 people we

realized that the new people joining

there's benefit of them having a space

to um to be next to others to get deeper

into the culture understand how what are

all the projects that are happening in

the company. We started the hops where

you can go into London and Warso and San

Francisco where you can work with others

in person and that's how we try to like

marry those two. If you are early in

your career you can you we we we try to

hire you in the hub so you can immerse

yourself in the company. If you are used

to remote work completely fine uh but

then if you want you can always come and

join us in the in the hub and that

worked really well. Currently we

continue like hiring very untraditional

backgrounds in some of the place of the

company and then fusing that with very

traditional backgrounds which can teach

the others and u and in sales for

example we we we we've done some of

those experiments too uh where that

combination worked really well

>> the lesson is you can really find talent

everywhere uh it just how hard and how

you look for them

>> and I think in Europe also people like

this was a interesting one in US people

are very keen and excited to to to work

and if you if you go for any social

event it's it's like you want to talk

about work and in Europe I didn't have

this feeling where it's like most people

don't want to do that it's like the

cultural piece is different but then you

do have the pockets of people that

actually strive it too they just don't

have the companies where they could do

that in so I feel like our our team from

Europe is is the most motivated and

passionate set of people that that that

that uh that we are lucky to have.

>> Yeah, I can attest to that given I've

met some of them. uh very hardcore, very

good work ethic for sure. And you have

also maintained a pretty flat or

structure um and have people own quite

laterally a lot of um responsibilities.

Can you talk about the rationale behind

that? And I guess there was also a no

title policy.

>> Yeah, so we removed titles a year ago

and then um and it's it's going well. it

still works and and I do think we you

know we I we we said we did it but I

thought a lot of AI companies kind of do

it too already with member of technical

staff being like a the usual piece you

have for engineering and then in a lot

of the go to market you are just go to

market not VP of sales or other roles so

I think it's it's actually a a pretty

common pattern but in our case we we we

had a small team approach where you have

extremely small amount of people usually

the five to 10 and um and we wanted to

make it very clear that every team we we

we create those teams. You have six

months to prove it. If it's proven, that

team will stay and continue working. But

it really is that the moment you join,

you can have any impact on the company.

So you can have any role in that team.

The tenure will not define your position

in the kierarchy. If you are smart and

quick and passionate, you can you can

you can elevate yourself very quickly,

which this this really this really

helped. And also it's a common um common

layer to the external world where

everybody looking at 11 labs knows that

we are the go to market team is go to

market team. There's no like uh

positioning to to the to the same

extent. What this allows us to do is I

think when we speak with a lot of our

partners with a lot of our customers um

they also know that uh that they are

getting the the best people um and

always and and we can also send people

to to different conferences different

events regardless of that that

positioning. Um I think the tricky thing

in the flat structure is not only

positives in the way we currently have

it's it's a set of leads effectively for

the subdivisions. So the research

creative work agents work go to market

self and salesled um and of course ops

um

only that's the layer of leads and then

under that there's pretty flat small

team approach across across the world.

Um but then you really want the leads to

be able to carry the complexity around

the team. So suggest things between one

team to another if they see that there's

something valuable between them

happening. Um so I think picking those

those people that can context switch

between is super important and then

letting the team fully focus on on that.

Um and then having which is uh which was

interesting learning where if you if you

put a person into all the Slack channels

and give them transparency they actually

get frequently distracted because then

they read all the messages. You can

still choose not to read them but they

still they still do. So you kind of need

to cut the access to a lot of those

pieces to force the attention and that

kind of works. All those small things

works work really well.

>> Maybe we can borrow some of that lesson

too.

Um let's switching gear a little bit. Uh

you're you're on the front line seeing a

lot of the creative work whether it's

from uh art, music or advertising that

are starting to adopt AI tools. And in

the beginning that was not the case.

there was a lot of resistance and now

we're just seeing the the adaptation and

the the welcoming of using more of the

generative AI tools including you know

AI audio um and you have done some

really smart things from the marketplace

payouts to like working with these

creative industries since day one

actually um I remember how much you

stressed like we have to find a way to

work with them and sort of um observing

sort of market shift over time so the

question is um how do you actually adapt

to these changes and find the ways um to

to work with the industry in the infancy

in the beginning and how did you

navigate some of the challenges in that

>> the um so I think the first piece is is

uh is actually spending time with the

industry and trying to understand what

are their priorities their incentives uh

of course it's sometimes tricky

sometimes you you then end up being

starruck we had a honor and pleasure to

work with Jared on on some of his

incredible work and uh and learn from

him on like what is important in the

like which parts of the production

process you can actually use AI which

ones you want to keep um where is it

actually helpful um uh uh and so so I

think that's the super important thees

across all the partnerships in the space

in our case we we try to figure out how

to do that on the on the voice space

which is of course with that technology

a how will the voice acting space look

like in the future and then too. Of

course, to be able to cut to all those

different use cases, you need such a big

array of different voices, different

languages, different accents, um

different styles. Um so, we launched

voice marketplace where you you could

create your voice and then uh share it

and when that voice is shared, you earn

money in the return. Today, we have

almost 10,000 voices. We paid $10

million back to the people in the

community. There's some crazy stories

from the voices. our one of our first

voices was a a deep Spanish voice and

the magic of the technology is that the

same voice now is available on all

different languages in the same way. So

it's 30 different languages at the time.

Now it's 70 but 30 languages at the time

and we had the Spanish voice join us and

it wasn't picking up on the Spain.

Nobody really liked it as much and then

it picked up in an English-sp speakaking

country that same voice because of that

deepness and now it's our top three

voice for all the use cases. So uh

hidden messages you can all register to

our voice marketplace and maybe earn

some money too. Um the so that's the the

I think the second important thing is

like figuring out how we can be part how

we can bring the industry together to

disrupt together rather than just to

disrupt and with labels I think I'm

still learning how to uh interact the so

we we've worked with labels uh the the

Merlin and Cobalt so four of majors to

bring their music into the music model

so we can do it in a licensed way. So

you can generate that and give

commercial rights so you're fully

protected. Uh that was a hard process.

It took us 18 months to figure out the

agreement that works. And in the end I

think the main thing uh was was adding

set of forcing functions or forcing

timings to um to find a a effectively a

trigger of like okay this is when we do

it and we either do it together or we do

it separately. and um and those forcing

functions really help add urgency. Then

we we needed to move that forcing

function a few times but but it still

worked to to a large extent to to go

after that. And then two there's of

course the you know the finding the

compromise wasn't wasn't wasn't wasn't

easy. Um but then in our case working

with the with the with the labels there

was um kind of protecting what they are

caring about and they of course also

care about how um how they continue

doing well by their members by their

artists that they work with. Um so we

would spend a lot of time working with

their members speaking about how we

think about technology what's going to

happen in the next couple of years and

that really helped. So just just

speaking through exactly the technology,

showing the examples and kind of

avoiding this initial knee-jerk reaction

that AI is bad has has been has been

tremendous.

>> And maybe tying back to the earlier

question as you are navigating like this

landscape. um how do you think about

like bringing the right talent that can

head and lead some of these functions

and these are mostly unknown territories

of how to navigate it like where have

you been seeing um success in bringing

the right people?

>> So here for the spaces that are kind of

completely new to us. So this and like

legal is another example. We would

always kind of bring at least one or two

people that were in that space that kind

of have interacted with the same parties

uh full-time in the past but then would

actually um uh uh adjust that with a lot

of consulting uh people that would help

us in a specific conversation. So in

this case in music we had uh music

lawyers that worked very closely with us

that consult across a few of them and

the good thing is that they know all the

players and they effectively were this

um uh bridging gap between between both

of us. So so we could speak the same

language and um and then that was that

was that was really helpful.

>> Yeah. And you have had um a very

specific taste for people that are risk

uh tolerant enough and also understand

the commercial business opportunities to

you know help guide the right chain of

actions in each of those domains. I

found that very fascinating

>> 100%. I mean legal I'm I don't know how

many of you are trying to find a first

legal council or have a number of those.

For us, this was the I think one of the

trickiest roles to hire for because you

are um hiring into the space you don't

know you know very little about. And

then uh and then we had a like the first

couple of legal people that that were

clearly not fed. So we separated us.

Then we hired a third person and that

person came from like a a number of

Fortune 500 companies and uh and they

never worked in startup space, never

worked in venture. And what resulted is

like everything every conversation was

pointing out the risks that we see. So

like anything we wanted to do was like

the number of risks that this could

carry. Um and it was really tricky to

work because we it's like you kind of

get risks but you got the risk advice of

like okay and this is where we should

draw the line. Uh but everything was bad

the decision and now we hired a person

working previously in the a number of

companies as a council and don't poach

them. They are amazing

and they understand the the the risk

equation a lot better where uh where

they are not only like a counterpart to

figuring out um what the risks are but

also like okay this is what other

companies do this is what we should

potentially do and then they're like a

true thought partner and the tremendous

change

>> for sure. Um 11 Labs uh started as more

of a creator brand um everywhere from

the individual creators to the the

creators that are building businesses

but now you have been having a lot of

success moving into enterprise um not

just started from the AI agent platform

but you know even with the the the text

to speech speech to text models um how

have you been navigating that transition

because that's one of the very common

place where you know a lot of really

great consumer creator brands fall down

But you have had so far a pretty smooth

transition.

>> So when we when we launched we had a lot

of early inbound where when when we

started the kind of the classic PLG a

lot of inbound from enterprise and I

remember speaking with A60Z team when

when they joined us where our initial

take was of course we want to be an

engineering company we don't want

salespeople we would like to reinvent

that and have like engineers do the

sales. Uh we we we did hire one

traditional salesperson and one

non-traditional salesperson like an

engineer and we told them like do sales

now and that really as you can imagine

didn't work out in this specific case.

Um but we learned our lesson uh and we

we now do invest in in a combination of

that. It's 80% sales 20% engineering. Uh

so still a little bit of that. Um but

this was like super important lever of

understanding who are the customers,

what they care about and working deeply

with them to to bring it back. Um and

then that kind of working with them was

kind of opening of what we need to

actually do on the product and research

side. Um uh Munjal from Hypocratic is is

is here. He was one of the uh the

earliest incredible use cases in the

healthcare space where they would create

effectively agents that would take

inbound calls that are calling the

hospitals to take and schedule

appointments and beyond that they would

do all the other parts of outbing to the

patients to remind them about taking

medicine or uh reminding them out the

appointment that's happening. And um and

to be able to do that that's suddenly

shifts from using a one foundational

model into combining the speech to text

the LM the text to speech to orchestrate

them together. Then the integrations you

need to build then you actually need to

deploy and they were one of the earliest

it was 2023 but then we've seen this

repeated pattern across uh across number

of uh other customers and customer

experience space and uh and many others

and um and we decided to invest more

into helping with the entire

orchestration. So instead of just doing

text to speech we can help combining our

research to make this whole whole

combination of that more fluid. But then

if you are thinking about enterprise you

do need to build um the combination of

knowledge base inside a system you need

to help deploy that with telefony

providers with twillio zip trunking like

how do you do that in a templatized and

easier easier way and then of course the

the biggest gap that's the most common

it's easy to do a demo but how do you

actually build it to production how do

you test how you version control how you

evaluate monitor over time fine-tune

over time based on the results and and

all of that is has been um a big big

part and underlying all of that and we

spoke a little bit with with Matt before

coming here the the foundation needs to

be there which is the security the

compliance serving serving the the

customers um across that will rely on

that infrastructure that's something

that we want to shine through OS 11 laps

where if you are using the software it's

going to always be reliable and always

um uh the 49ths or 59s hopefully one day

will be will be there which is tricky in

AI space uh but the the that's the

that's the goal of course the the

difference between um the the one

obvious difference between PLG and sales

is the the cycle to work through and

identify the right customers is much

longer and um and I think that's where

eagerness from our internal team was was

was was interesting to observe where you

had a lot of people that didn't work in

a enterprise setting and then you had

other side of the company that did and

the side that didn't was very skeptic

about going enterprise and like kind of

waiting the six months or 12 months to

results and in the early days we needed

to shield them from that information and

like trust us we'll do this and it will

work. Uh but they were very skeptic and

then of course after 12 months it it it

worked out but that was probably the

hardest culturally of how you kind of

still keep everyone jumping on the same

on the same train.

>> That's exactly right. Um a lot of

companies actually at least I observed

sort of slowed down after start adopting

uh more of the enterprise sort of

product launching and like building for

the customers request um that started to

Thank you so much. um to delay sort of

the the product launches. um is that

something you're seeing or is there

still like a good balance of like we

still want to be able to put out demos

and PC's and um early teasers quickly

but at the same time we'll get to you

know deliver a very robust and reliable

product. So there are two parts. The the

first part is um so we have a like a

difference on the team structure and

then we have a difference on the kind of

external product structure. On the

external product structure

we we want to ship very quickly. But of

course if you are shipping to enterprise

you you want to make sure that it's

stable and reliable. So we delineate

very clearly what's alpha, what's not

alpha. Um and then we go through that

transition through through that period.

Um and then as we work with the

customers they can and then our partners

they can decide whether whether they

want the access to alpha in the first

place and when they do they that's

clearly shown that this is an alpha

product that might not be as stable and

so they get a choice and I think that

choice has been the the most important

lever like do you do do you want it or

not and and some are are are um

incredible on doing that that um

innovation and and and and showing some

of their work or experimenting with that

work. uh Deutsche Telecom with John here

is is is is creating some of the

incredible new podcast experiences and

that came from like testing early models

of turning a a text a text into like a

more notebook LM style of a podcast with

incredible voices that you can select

for German speaking voices, English

speaking voices that um that that sound

good. Um and then there's a second which

is team structure piece and that's

something that we didn't do until until

later when we had uh more than hundred

of us is that we delineate inside a

company products that are pre-product

market fit and postp product market fit.

Um on the postproduct market fit you are

working for the long term. You test and

evaluate a lot before you um you only

deploy when when that's that's that's

truly ready. the pre-product market fit.

Your mission is is to ship until you

think we've hit the product market fit.

And usually we give the six months

period of like proving it out. If not,

we kill the product and we've killed

product in the past uh this way. But

that's like the the main important piece

of like okay until we know there's a big

potential user base, we we we will

continue iterating.

>> I've uh been able to observe some of

those uh I guess hard decisions in the

moment, but it's the right decision

later on to to let go some of products.

Um this is one of my favorite questions.

Uh my partner Martin Casado always say

companies go through three phases. There

is a product phase, there's sales phase

and there's a scaling phase. And given

you have been through some of those

phases, what has been the hardest

transition for you as a CEO?

>> There's a lot of a lot of many ones. Of

course, I have my my co-founder next to

me across each of those which is the I

know him for 15 years. He's my best

friend since high school. say like the

the most luck to to to have uh that

combination of course uh you Jennifer

and all the all the partners to help us

through those transitions which is which

has been incredible. Um but I think the

the the the recent like recent

realization was when we we are now 350

people company and uh and of course that

means our go to market team and the

incentive structure around that has

evolved pretty pretty strongly and what

wasn't clear to me and now in hindsight

it's obvious is that the um in early

days everybody would just operate on a

passion basis. They would just operate

what they think is best for the company.

As our go to market team enlarged, we

realized that the incentive structure

really matters if you are building that

machine. And um and that transition

where you shift from

from from a lot of a lot of the people

that are helping create that machine are

part of that machine. Those incentive

structures will eventually drive the

behaviors which might be slightly

different to what you had in mind if you

don't make it extremely clear. And in

some ways the the quat the commissions

are a effectively a lagging indicator of

strategy and then um and then strategy

um uh is kind of leading of what will

happen in the future. So you need to

find a way to resolve those two together

where you want to make sure the quad end

commissions and the strategy that you

want to drive are closer together and

they and the the kind of the disparity

as close as possible. And uh so so here

the for me the biggest realization was

that we are becoming a bigger company

because there are clear behaviors that

happen based on the commissions. And

then two to actually resolve those we

need to be very upfront in terms of um

of uh of making it explicit that

sometimes even if commissions are just

this and you think it's a wrong thing

come back to us let's speak about it and

just course. So now we are explicit with

all our sales teams that if they are

seeing a deal that let's say might be

competitive in nature and our pricing

table would suggest that they can go

very low and earn higher commission but

they think it's wrong it's better to

come to us. We are happy to still grant

commission but kill the deal and um and

and go. We had this case recently where

one of our foundational level competitor

came to us wanting to license our models

for demos and um and of course the

incentive would suggest that you should

sell to them but luckily luckily we

didn't.

>> Yeah,

>> you granted commission though.

>> Yeah, in early days you can definitely

>> and adjusted that now it's in the policy

so you cannot sell to the foundational

model companies.

>> So it's clear clear to to all the

internally.

>> Um that was incredible Maddie. Thank you

so much for for sharing all the lessons

and learnings with us. Let's give a

round of applause to to Maddie.

>> Thank you.

[Applause]