June 2025 Summaries

ElevenLabs and Cartesia are two AI audio platforms compared based on features such as voice quality, pricing, and language support. ElevenLabs stands out with its extensive library of over 4,000 voices and support for more than 70 languages, offering unparalleled voice realism and a wide range of functionalities, including AI dubbing into 29 languages, professional voice cloning, and additional products like ElevenReader and Speech to Speech conversion. In contrast, Cartesia offers a more limited selection, supporting only 15 languages and around 130 preset voices, with basic instant voice cloning capabilities. Both platforms provide API access and pricing tiers catering to various user needs, but Cartesia is generally more cost-effective for simpler projects. While ElevenLabs is preferred for high-quality, multi-lingual, and long-form content projects, Cartesia may be suitable for users with less demanding requirements.

The text provides a detailed comparison of PlayHT with other text-to-speech (TTS) platforms, highlighting their features, capabilities, and performance. It examines PlayHT and its alternatives, such as ElevenLabs, Speechify, Microsoft, Google, Amazon Polly, and OpenAI, focusing on aspects like voice quality, emotional delivery, language support, and integration possibilities. ElevenLabs is noted for its superior voice realism and emotional expressiveness, outperforming PlayHT in a survey and offering features like voice cloning and AI dubbing. PlayHT, however, boasts a vast voice library and is praised for its ease of use and broad language support. Both platforms offer tiered pricing models with free trials, catering to a range of users from hobbyists to enterprises, while also ensuring ethical AI use and user privacy. The text emphasizes the versatility of these TTS services across industries such as content creation, e-learning, and digital media, and outlines their customization options, integration capabilities, and user support resources.

MOST, an AI-powered audio guide premiered at the Jewish Culture Festival in Krakow, offers an immersive narrative experience that brings Jewish heritage to life through AI-generated voices. Developed by sonic systems artist Andrew Melchior, the guide uses ElevenLabs voice technology to create a location-based storytelling experience that combines AI voice synthesis, archival materials, and personal testimonies to reveal Jewish life in interwar Poland. As visitors navigate the Kazimierz district, they encounter multilingual, adaptive voices that connect personal experiences with collective memory, facilitated by a female Polish narrator and a fictional Jewish male voice inspired by historical sources. This initiative, supported by ElevenLabs Impact, reflects a commitment to cultural and civic uses of AI, aiming to preserve and animate Krakow's Jewish history through innovative digital means.

Adobe Captivate has integrated ElevenLabs' AI voice technology to enhance eLearning courses by providing lifelike narration directly within the app, allowing course creators to convert closed captions into natural-sounding audio using over 150 AI voices in 10 languages. This integration aims to make voice a core element of the creative process, improving comprehension and engagement while streamlining workflows by eliminating the need for third-party solutions. Early adopters of this technology report faster production cycles and more dynamic content, which lead to highly personalized audio experiences for learners. ElevenLabs' technology is positioned as a significant innovation in the learning platform, enabling educators and course designers to focus on delivering impactful and accessible learning experiences globally.

Funding Societies, Southeast Asia's largest SME digital financing platform, partnered with ElevenLabs to automate their sales outreach using Conversational AI with custom multilingual voice agents. This collaboration allows Funding Societies to make over 1,000 daily outbound calls across different countries and languages, with AI handling initial interactions and human agents engaging only after qualification. The AI uses voice cloning technology to maintain brand consistency by replicating the tone and professionalism of live agents. ElevenLabs was chosen for its expressive, realistic speech and rapid integration, ensuring low latency and high-quality interactions. The AI system not only handles sales calls but also expands into payment reminders and post-loan engagement, creating a scalable communication framework that allows human teams to focus on more complex tasks.

Voice Design v3 from ElevenLabs is an advanced tool for creators, businesses, and developers to create unique AI voices with ease by describing their desired voice and instantly receiving three options. This version enhances the speed and control of voice creation and offers high precision audio quality and a smarter prompting engine capable of handling nuanced cues without artifacts. It supports two design modes: Realistic Voice Design for lifelike performances and Character Voice Design for more imaginative creations. Users can create a voice in three steps: pick a concept, prompt with intent, and save and deploy the chosen voice, with applications spanning game development, audiobooks, podcasts, localization, and accessibility. Voice Design v3 is accessible via the ElevenLabs dashboard and complements the existing Professional Voice Clones by providing a flexible solution for generating new voice concepts.

ElevenLabs Voice Library is a platform where individuals can monetize their voices by uploading high-quality recordings, allowing their voice clones to be used for various commercial and creative purposes. The system operates on a royalty basis, paying contributors based on the character count generated with their voice, with a default rate of approximately $0.03 per 1,000 characters. Contributors maintain control over the use of their voice, including pricing and usage moderation, and can enhance discoverability through tagging by accent, age, or tone. Earnings are processed weekly via Stripe Connect once a balance of $10 is reached, and voices with unique qualities may command higher rates. The platform emphasizes ease of entry, requiring only clear audio recordings and offering opportunities for passive income, as voices can continue to earn royalties over time without further input from the creator.

ElevenLabs has launched a mobile app for iOS and Android, offering its advanced AI voice tools directly on smartphones, enabling users to create ultra-realistic voiceovers using the latest Eleven v3 Text to Speech model. This app caters to content creators, marketers, educators, and voice artists who require a mobile solution for integrating AI voice tools into their creative workflows, allowing for greater flexibility and mobility. The app includes features such as selecting favorite voices, exporting audio clips for use in various editing tools, and syncing with web accounts, all while leveraging the expressive capabilities of the Eleven v3 model to deliver nuanced and lifelike speech. With this launch, ElevenLabs aims to make content accessible in any voice and language globally, enhancing the creative possibilities for users on the go.

ElevenLabs is now providing the voice technology for Cisco's Webex AI Agent, aiming to enhance customer support by utilizing natural, expressive voice interactions. This collaboration addresses significant gaps in customer service, as highlighted by Cisco's research, which found that a large percentage of customers are dissatisfied with their service experiences. Traditional chatbots have often failed due to their rigidity and inability to understand natural language, but the Webex AI Agent leverages large language models to offer human-like, adaptable experiences. ElevenLabs' technology enables the AI to respond to emotional cues and deliver interactions that mimic real human conversation, thus improving customer satisfaction. This partnership combines advanced voice technology with Cisco's scalable support and design tools, allowing for seamless integration with enterprise systems like CRM and ERP, and aims to transform customer support operations globally by delivering more intuitive and effective AI-powered customer engagements.

11.ai is an alpha-stage voice-first AI assistant introduced by ElevenLabs, designed to integrate with existing tools via the Model Context Protocol (MCP) to perform meaningful actions beyond basic question-answering. This assistant aims to enhance productivity by allowing users to execute tasks such as planning, research, project management, and team communication through voice commands, leveraging integrations with platforms like Linear, Slack, and Notion. Built on ElevenLabs' Conversational AI technology, 11.ai supports real-time voice interaction with minimal latency, integrates with a variety of APIs, and offers customizable voice options, creating a personalized user experience. The platform is currently available for free during its experimental phase, encouraging user feedback to refine its capabilities and expand its integration options, thereby showcasing the potential of voice-first productivity in modern workflows.

Augie, a platform aimed at democratizing video creation, enables professionals to produce marketing videos without traditional production teams by turning scripts into polished content using pre-licensed assets from Getty Images. The integration of ElevenLabs' AI voices allows marketers to create videos with natural-sounding narration, eliminating the need for personal involvement in video production. Augie, one of ElevenLabs' earliest API partners, chose their text-to-speech model for its voice quality and ease of integration, leading to a significant portion of videos on the platform relying on AI-generated scripts and voices. This approach empowers creators to produce full videos without cameras or microphones, expanding their creative possibilities and global reach. The platform is currently enhancing its offerings with multilingual support and additional features, demonstrating how AI can scale creative output efficiently and serve as a model for product-led growth in the video production industry.

PERSO.ai has partnered with ElevenLabs and ESTsoft to streamline the process of video localization by integrating natural voiceovers and frame-accurate lip-sync technology, facilitating the creation of culturally precise multilingual content. This collaboration combines ESTsoft's dubbing tools and Cultural Intelligence Engine with ElevenLabs' multilingual voice AI, enabling creators to produce content that resonates with local audiences in over 30 languages. The new system supports platforms like YouTube, TikTok, and Google Drive without requiring technical setup and is already being used to localize over 50,000 minutes of content monthly. By significantly reducing turnaround times and production costs, this innovative approach allows video creators, training teams, and marketing departments to focus more on storytelling while meeting the demands of the growing creator economy.

ElevenLabs is hosting an online Conversational Agent Hackathon on July 2, 2025, to celebrate the creation of 1 million agents, offering prizes over $20,000. Participants will have two hours to build a voice agent on the ElevenLabs platform and can submit their creations for both the main ElevenLabs prize and additional partner track bonuses from companies like Exa, Notion, and n8n. The event will be supported via Discord, where technical support and general announcements will be available. The hackathon aims to encourage innovation in various fields such as customer service, education, and healthcare, by leveraging conversational AI technology. Participants are encouraged to register, build their agents during the event, and submit a video demonstration of their work within 24 hours post-event, with winners announced a week later.

Eleven v3 Audio Tags introduce a revolutionary way to emulate accents seamlessly within AI-generated speech, enabling transitions between various accents such as American, British, French, and Australian mid-sentence, script, or character. This feature allows creators to produce dynamic and culturally enriched voice performances without the need for separate voice models or manual retakes. Accent emulation in AI speech refers to modifying pronunciation and rhythm to reflect different dialects while maintaining the original words, offering creative and cultural range for content localization, character identity, and geographically grounded dialogue. Through the use of tags like [French accent] or [Southern US accent], users can direct the model to deliver accents contextually, enhancing experiences like audiobooks, gaming characters, and product demos. Although Professional Voice Clones are not yet fully optimized for Eleven v3, Instant Voice Clones or designed voices can be used to explore these features during the research preview stage.

The text discusses the capabilities of Eleven v3 Audio Tags, which offer fine-grained control over the timing, rhythm, and emphasis of AI-generated speech, transforming flat delivery into dynamic performances. By using specific tags such as [pause], [rushed], and [drawn out], users can manipulate the pacing and emotional impact of spoken lines, making them feel dramatic, casual, tense, or comedic. This allows for precise direction of speech delivery, affecting how lines are interpreted based on timing and intent rather than just word choice. The text also notes that while Professional Voice Clones are not yet fully optimized for Eleven v3, alternative options like Instant Voice Clones are available for projects requiring advanced features. The piece emphasizes that Eleven v3 turns scripts into scores, enabling creators to manage delivery with precision, thus enhancing the believability and engagement of AI-generated audio content.

Voice cloning, an emerging technology powered by artificial intelligence, enables the replication of a person's unique vocal characteristics, allowing for the generation of speech that closely mirrors the original speaker's tone, pace, and style. This process involves collecting diverse voice samples, training machine learning models to recognize vocal patterns, and synthesizing new speech that sounds natural and personalized. Voice cloning has practical applications, particularly for individuals who have lost their ability to speak due to medical conditions, as it preserves their vocal identity. Additionally, it offers benefits to creators and voice actors by allowing them to license their voices for use in various digital formats without the need for repeated recordings. Despite its advantages, voice cloning carries potential risks, such as misuse for impersonation or misinformation, prompting companies like ElevenLabs to implement safeguards, including identity verification and moderation tools.

Eleven v3 Audio Tags enable the creation of dynamic multi-character dialogues, allowing a single AI voice model to play multiple roles in a scene with different styles, tones, and rhythms. This technology facilitates natural dialogue by incorporating overlapping voices, interruptions, and emotional shifts, which were traditionally managed by multiple speakers and recordings. Through the use of tags like [interrupting], [overlapping], and [laughs], users can script scenes where voices interact fluidly, enhancing the realism and engagement of conversations. While Professional Voice Clones are not yet fully optimized for Eleven v3, Instant Voice Clones or designed voices are recommended for projects needing these advanced features. This innovation is particularly beneficial for storytellers, game writers, and interactive designers, as it simplifies the production of complex, orchestrated performances without significant overhead.

StudyLabAI is revolutionizing personalized education by incorporating ElevenLabs' voice AI technology to provide real-time, multilingual tutoring with the "Tutor Me Companion," which uses Text to Speech and Conversational AI to facilitate interactive, human-like learning experiences. With the help of an ElevenLabs Grant, StudyLabAI quickly integrated this technology, allowing students to engage in natural-sounding dialogues across various subjects in their native languages. The seamless integration process led to significant business impacts, including a 35% increase in premium subscriptions and a 45% improvement in user retention, with the full setup completed in under a week. Voice AI has become central to StudyLabAI's strategy to scale and innovate, as they continue to shape the future of AI-powered learning platforms, making high-quality education more accessible worldwide.

Eleven v3 Audio Tags introduce a new level of narrative intelligence in AI speech, enabling dynamic storytelling by using tags such as [pause], [awe], and [dramatic tone] to manipulate the emotional rhythm and structural flow of narration. This technology allows AI to not only synthesize voice but also direct storytelling by understanding when to convey suspense, irony, or reflection, thereby making the narration feel more personal and engaging. These tags are versatile and applicable to a wide range of formats, from documentaries to internal monologues, guiding attention and setting moods. While Professional Voice Clones (PVCs) are not yet fully optimized for Eleven v3, Instant Voice Clones (IVCs) or designed voices are recommended for projects utilizing this model, with further PVC improvements anticipated in the future. This development represents a significant advancement in voice storytelling, offering creators the ability to design the pace, tone, and emotional structure of a scene directly from a text editor.

Eleven v3 Audio Tags enhance AI speech by infusing it with emotional nuances such as tension, warmth, hesitation, and relief, making spoken content more relatable, dynamic, and human-like. This capability allows for context-aware performances where emotional context influences how a character reacts to situations, guiding the emotional state mid-delivery with bracketed cues like [sigh], [excited], or [tired]. These emotional tags help control pacing, tone, and atmosphere in narration, dialogue, and UI feedback, enabling creators to engage audiences without re-recording or rewriting. However, while Professional Voice Clones are not yet fully optimized for Eleven v3, Instant Voice Clones or designed voices are recommended for projects utilizing these features during the research preview stage.

ElevenLabs' Eleven v3 introduces Audio Tags, a feature in their new Text to Speech model that enhances character performance by allowing users to control tone, emotion, and pacing in speech. This tool enables precise direction over vocal identity, making it possible to switch accents, dialects, and archetypes like villains or narrators within a script without changing the underlying text or voice. This flexibility is ideal for applications such as animation, games, and interactive fiction, where character voice is crucial. Audio Tags allow for dynamic vocal changes and contextual shifts, transforming text into a performance that matches the desired persona. While Professional Voice Clones are not yet fully optimized in v3, users are encouraged to use Instant Voice Clones or designed voices during this research preview.

KPN, the largest telecom provider in the Netherlands, has partnered with ElevenLabs, a leader in voice AI technology, to integrate advanced AI audio into Dutch services. This strategic partnership, supported by a strategic investment from KPN Ventures, aims to enhance real-world applications like voice-accessible content and automated customer support, offering natural, human-like audio experiences to consumers and businesses. The collaboration focuses on making content more accessible via voice and improving automation and customer interactions, laying the foundation for more personalized and seamless voice-driven experiences within KPN's ecosystem. Both companies are committed to accelerating the adoption of voice AI in the Dutch market, emphasizing intuitive, natural, and human voice tools that prioritize quality, privacy, and usability.

ElevenLabs offers a suite of AI-driven tools designed to streamline the process of creating professional voiceovers for YouTube videos, enabling creators to generate natural-sounding audio quickly and efficiently. By utilizing Text-to-Speech and Voice Cloning technologies, users can create lifelike voiceovers without the need for extensive recording sessions, allowing for rapid content production while maintaining high quality. ElevenLabs provides access to a vast library of over 5,000 voices in various tones, accents, and languages, alongside features such as a dedicated Studio for long-form content, Voice Isolator for audio clarity, and Dubbing Studio for multilingual localization. These tools allow creators to tailor their audio content to specific audiences and maintain brand consistency across different languages, all within a browser-based platform that eliminates the need for additional software or hardware.

Eleven v3 Audio Tags, a feature of the new Eleven v3 (alpha) Text to Speech model, enable users to control AI speech by adjusting tone, emotion, and pacing to match real-world contexts, providing situational awareness to the AI. These tags, which appear as words in square brackets, serve as performance cues that allow the AI to adapt its delivery mid-sentence, transforming narration into a performance that reflects emotional beats or situational shifts. This innovation is particularly valuable in dynamic or high-context scenes, such as sports commentaries or suspenseful audiobooks, where tags like [EXCITED], [WHISPERING], or [SHOUTING] can dramatically influence how content is perceived. The model's ability to shift tone mid-line and handle interruptions without rewriting scripts offers a new layer of creativity for voice designers, game developers, and storytellers, though Professional Voice Clones are not yet fully optimized for this version.

ElevenLabs has released Eleven v3, an alpha research preview of a new AI voice model that introduces Audio Tags to enhance control over emotion, pacing, and sound effects in text-to-speech applications. These tags, which are words enclosed in square brackets, allow users to direct the AI voice to express emotions, delivery styles, and even nonverbal cues like pauses and tone, thus elevating the expressiveness of generated speech. This feature is particularly useful for producing immersive audiobooks, interactive characters, and dialogue-driven media, offering precise control over audio delivery. Despite Professional Voice Clones (PVCs) not being fully optimized for Eleven v3, Instant Voice Clones (IVCs) or designed voices can be utilized to explore v3's features. Available in the ElevenLabs UI and through a public API, Eleven v3 is currently offered at a discounted rate, encouraging experimentation with its enhanced capabilities.

ElevenLabs offers a comprehensive guide to creating professional-grade voice clones using their Text to Speech technology, emphasizing the importance of high-quality input data and precise prompts for optimal results. The process involves starting with pristine recordings in quiet environments using suitable equipment, capturing expressive and varied speech, and ensuring a clean dataset devoid of flaws such as filler words and inconsistent recording conditions. The guide highlights the significance of maintaining consistency in recording conditions, providing the right amount of training data tailored to the intended use case, and fine-tuning settings for stability and similarity. It also suggests stress-testing voice clones in real scenarios to evaluate their performance across different contexts and advises on managing voice clone libraries effectively through naming conventions, version control, and metadata documentation. ElevenLabs encourages users to experiment and iterate, offering both a free tier and upgrade options for additional features like voice mixing and multilingual cloning.

Xaia, a clinical assistant designed to enhance patient care, has successfully integrated ElevenLabs' advanced Speech-to-Text (STT) and Text-to-Speech (TTS) technologies to streamline medical workflows and improve clinical outcomes. The initial STT model Xaia used was flawed, generating inaccurate transcriptions and missing crucial non-verbal cues, which are essential in clinical settings. By adopting ElevenLabs' Scribe, Xaia significantly reduced transcription errors and enhanced the contextual accuracy of patient interactions, capturing important sounds like laughter and coughing. This technological shift has halved the documentation time for clinicians and provided more reliable mental health support for patients, resulting in better patient outcomes and more efficient clinical operations.

Eleven v3 (alpha) is a new Text to Speech model unveiled by ElevenLabs, offering unprecedented expressiveness and control in speech generation across 70+ languages. It introduces features such as multi-speaker dialogue, audio tags for controlling tone and emotion, and a new Text to Dialogue API for creating natural-sounding conversations. While it requires more prompt engineering compared to previous models, it significantly enhances expressiveness with capabilities like sighing, whispering, and laughing in speech. Designed for applications like videos and audiobooks, Eleven v3 is not recommended for real-time or conversational use cases due to its higher latency and need for optimization. A real-time version and better support for Professional Voice Clones are in development, with current availability through ElevenLabs' website and API, and an 80% discount offered until the end of June 2025 for self-serve users.

Text-to-Speech (TTS) technology has evolved significantly from its utilitarian roots, now offering fast, emotional, and lifelike voice outputs that mimic human speech patterns, adjust tone, and support multiple languages. This advancement opens diverse possibilities for creators in various domains, such as narrating blogs, localizing videos, or adding speech to apps, facilitating faster content creation without sacrificing nuance or brand consistency. ElevenLabs emerges as a leading tool in this arena, allowing users to turn text into expressive, scalable speech with features like voice cloning, emotional nuance, and multilingual support, all accessible through a simple UI or powerful API. The platform emphasizes ease of use, enabling users to transform written content into natural-sounding audio by selecting from an extensive voice library, adjusting speech styles, and integrating the output into multiple formats such as videos, podcasts, and apps. TTS tools like ElevenLabs enhance accessibility, audience engagement, and content scalability, offering both free and paid plans to accommodate individual and enterprise needs, with additional resources available to help users leverage the full potential of TTS capabilities.

ElevenLabs Soundboard creator, known as SB1, is a browser-based tool that leverages the ElevenLabs Text-to-Sound Effects model to enable users to generate and manipulate custom sound effects using simple text prompts. Users can describe the desired sounds in plain language, and the AI produces four variations for each prompt, allowing users to select and assign sounds to pads for triggering. This tool streamlines beat-making by eliminating the need for extensive searches through sample libraries, as it quickly brings sounds to life based on user descriptions, offering flexibility and creativity without relying on pre-existing libraries. Designed for both novices and experienced creators, the sound effects generated are royalty-free, making them suitable for various creative projects. The platform encourages experimentation by allowing users to tweak prompts for different versions of sounds, build beats using unconventional elements, and layer or loop sounds to create dynamic audio compositions.

ElevenLabs' SB1 Soundboard is a browser-based tool designed for creating fast, AI-generated audio playback without the need for uploads or editing, optimized for initial setup and experimentation but less so for live performances. To enhance its live performance utility, it can be paired with a MIDI controller, which provides tactile control and instant sound triggering, thereby reducing delays associated with mouse clicks. The connection process involves using a MIDI-to-keystroke translator software that maps MIDI inputs to soundboard functions, allowing users to trigger sounds seamlessly during live sessions. This combination of SB1's AI-driven sound generation and the intuitive physical interaction offered by MIDI controllers aims to keep creators in their creative flow by minimizing distractions and enhancing responsiveness.

Vibe Draw is an innovative voice-first creative tool developed as a weekend project by Ryan Morrison, combining ElevenLabs' voice AI with FLUX Kontext for voice-powered image creation. This tool allows users to create and manipulate images by simply describing them out loud, leveraging FLUX Kontext's ability to generate and edit images based on spoken prompts. Vibe Draw utilizes various technologies, including the Web Speech API for speech recognition and ElevenLabs' text-to-speech API for responsive voice interactions, all running client-side for lightweight functionality, though it recommends server-side handling for production security. The system tackles challenges like natural language understanding and contextual awareness to distinguish between new creations and edits, ensuring seamless user experience with an audio queue system to manage responses. The project demonstrates the potential of conversational AI in visual creativity, removing barriers between imagination and execution, and opens possibilities for new capabilities like multimodal input and collaborative sessions.

AI-powered Text-to-Speech (TTS) technology is transforming digital accessibility by converting written text into spoken words, thus removing barriers for individuals with visual impairments, reading difficulties, or language barriers. Modern TTS systems, like those from ElevenLabs, use advanced machine learning and speech synthesis to create natural-sounding voices that mimic human speech, allowing users to engage with digital content more intuitively. This technology not only supports multiple languages and accents but also offers customization options for voice speed, pitch, and inflection, making it useful for a wide range of applications from educational tools to enterprise solutions. By integrating TTS through APIs, developers can enhance accessibility and compliance with web content guidelines, ensuring that digital environments are more inclusive and engaging for all users.