Speech AI Companies for Voice Recognition and Text to Speech

Suzhou Shi, Jiangsu Sheng

AISpeech Co., Ltd. is a prominent Chinese company focused on artificial intelligence speech technology. Founded by graduates from Cambridge University, the company is headquartered in Suzhou, China. AISpeech specializes in speech recognition, synthesis, semantic comprehension, and intelligent dialog, particularly for conversational AI platforms tailored for enterprise applications. The company offers a wide range of AI-powered products and solutions. These include meeting and office solutions like AI office laptops, advanced microphones, and conference speakers designed to enhance collaboration. In the automotive sector, AISpeech provides voice technology SDKs and in-car voice assistants. Additionally, it develops voice interaction technology for consumer electronics, enabling smart upgrades for various appliances. AISpeech's core technologies encompass speech recognition, synthesis, and dialog management, supporting diverse applications across multiple industries.

1702007

Altered

London, England

Altered Ltd, operating as Altered.ai, is a London-based startup founded in 2018 that focuses on digital audio content creation using artificial intelligence. The company aims to democratize audio production by allowing users to alter vocal identity, accent, and performance style while maintaining tempo and tonality. Altered offers a voice content creation platform called Altered Studio, which integrates various voice AI technologies. Key features include speech-to-speech voice morphing, voice changing for media production and real-time communication, high-quality text-to-speech, voice cloning, and tools for audio editing. The platform is designed for users in the media production and communication industries, including video game developers, voice application creators, and brands needing voiceovers. With a small team and significant funding, Altered is positioned to make professional-grade voice performance accessible to a wider audience.

AssemblyAI

San Francisco, California

AssemblyAI is a San Francisco-based applied AI company founded in 2017. It specializes in providing Speech-to-Text and Speech Understanding APIs that help developers and enterprises transcribe, analyze, and create voice-enabled applications. The company aims to democratize access to advanced AI models, allowing developers to build voice AI products more efficiently. AssemblyAI offers a unified API platform with three main services: high-accuracy Speech-to-Text for converting audio and video into text, ultra-low latency streaming Speech-to-Text for real-time transcription, and advanced Speech Understanding that extracts insights from audio data. Their technology is utilized across various industries, including media, customer service, healthcare, and technology, serving a global audience of developers and enterprises. AssemblyAI is trusted by over 10,000 companies, including notable clients like Zoom, Delta Airlines, and Spotify, positioning itself as a key infrastructure partner in the Voice AI space.

952017

AudioShake

San Francisco, California

AudioShake is a San Francisco-based AI company founded in 2020 by Jessica Powell and Luke Miner. The company specializes in advanced audio separation technology, enabling the "unmixing" of audio recordings into individual components, known as stems. This technology allows users to isolate specific sounds such as vocals, instruments, and dialogue, making it applicable in the music, film, and media industries. AudioShake's flagship product is a cloud-based AI platform that disassembles audio tracks, providing capabilities for music remixing, dialogue isolation, and ambient sound extraction. The platform also supports stem creation for monetization and offers API and SDK solutions for enterprise integration. With a focus on high fidelity and precision, AudioShake serves over 40 enterprise clients, including major names in the music and film industries. The company has secured significant funding, totaling $19 million, to further its mission of enhancing audio creativity and accessibility.

412020

Audo AI

New York, New York

Audo AI is bringing state of the art artificial intelligence to the media processing industry through simple-to-use, powerful, developer-facing APIs. Audo AI APIs will radically improve every stage in digital media processing from capturing and enhancing to editing and distribution empowering businesses to create amazing new experiences.

12020

BeyondWords

London

BeyondWords is an AI voice and audio publishing platform founded in 2016 and based in London, United Kingdom. The company provides a comprehensive audio solution for modern publishers, enabling them to convert text content into spoken-word audio using synthetic voices. BeyondWords automates the entire audio content lifecycle, including production, distribution, analytics, and monetization, making it easier for newsrooms, businesses, and independent writers to integrate text-to-speech capabilities into their content. The platform features an extensive AI voice library with over 550 to 720+ synthetic voices across 64 to 140+ language locales. Users can create custom voices, utilize a text-to-speech editor, and benefit from seamless CMS integrations. BeyondWords also offers distribution tools, performance analytics, and monetization options through advertising or subscriptions. The service caters to a diverse range of customers, including news publishers, independent writers, and large organizations in various sectors. Pricing includes a free plan and several paid tiers to suit different needs.

102018

Cerence

Burlington, Massachusetts

Cerence AI is a leading provider of conversational artificial intelligence, focusing on voice and interaction technologies for the automotive industry. Founded in 2019 and headquartered in Burlington, Massachusetts, the company has embedded its solutions in over 525 million vehicles worldwide. Cerence specializes in creating natural and safe interactions for drivers and passengers through its proprietary automotive-specific language models, known as CaLLMâ„¢. The company offers a comprehensive software platform that enables the development of branded virtual assistants for various functions, including control, navigation, communication, and media. Key products include advanced speech recognition, natural language understanding, gaze tracking, and gesture recognition. Cerence serves a diverse range of customers, including leading automotive manufacturers, Tier-One suppliers, and technology companies, and its technology extends beyond vehicles to applications in planes, tractors, and cruise ships. With a global presence and a workforce of 1,001 to 5,000 employees, Cerence is positioned as a vital player in enhancing the voice-powered experience in mobility.

13002019

Cleanvoice AI

Bucharest, Bucharest

Cleanvoice AI is an artificial intelligence-powered audio editing tool based in Bucharest, Romania. It specializes in automatically removing unwanted elements from audio and video files, such as filler words, mouth sounds, background noise, and long silences. The platform is designed for podcasters and content creators, enhancing audio clarity without the need for manual editing. The tool features a user-friendly web-based interface that allows users to upload their audio or video files easily. It supports various formats and sample rates, providing quick processing and delivering cleaned files via download or email. Recent updates include a "Studio Sound" feature that improves audio quality by restoring frequencies and applying auto-equalization. Cleanvoice AI is continuously developing additional features to enhance its offerings.

Deepdub

Plano, Texas

Deepdub is an enterprise voice infrastructure company founded in 2019, based in Tel Aviv, Israel, with additional offices in Plano, Texas, and Los Angeles, California. The company specializes in AI-powered dubbing and audiovisual localization, helping content creators and studios overcome language barriers while maintaining the original voices and emotional nuances of actors. Deepdub offers an end-to-end platform that automates translation, adaptation, voice cloning, and mixing, significantly reducing dubbing timelines. Their services include Deepdub GO, a self-service virtual AI dubbing studio, and a managed "White Glove" service for comprehensive post-production support. The company also features proprietary technologies for zero-shot voice cloning, voice-to-voice translation, and emotion-based text-to-speech. Deepdub primarily serves the B2B Media & Entertainment sector, including film studios, television networks, and online platforms, as well as corporate sectors, eLearning platforms, gaming companies, and advertising firms. The company aims to bridge language and cultural gaps in entertainment through high-fidelity AI localization.

2402019

Deepgram

San Francisco, California

Deepgram is a foundational AI company based in San Francisco, California, specializing in real-time voice AI technology. Founded in 2015, it aims to enhance human-to-machine interactions with high accuracy and low latency. The company supports over 200,000 developers and more than 1,300 organizations, establishing itself as a leader in the Voice AI sector. Deepgram offers a range of advanced voice AI solutions through a flexible API platform. Its key products include a Speech-to-Text API for transcribing audio, a Text-to-Speech service for generating natural-sounding audio, and Audio Intelligence features for sentiment analysis and topic detection. Additionally, the Voice Agent API enables the creation of autonomous voice agents, while Deepgram Saga streamlines development workflows with natural speech control. The company has processed over 50,000 years of audio, showcasing its reliability and scale for enterprise clients across various industries, including finance, government, and software.

3702015

Descript

San Francisco, California, United States

Descript is an AI-powered video and audio editing platform founded in 2017 and based in San Francisco, California. The company provides innovative tools that allow users to edit media content by typing text, making professional-grade production accessible to creators, marketers, and enterprises. With around 183 employees, Descript has raised $100 million in venture capital, including a $50 million Series C round in 2022. Descript offers a comprehensive suite of cloud-based tools for recording, editing, transcribing, and sharing video and audio content. Key features include text-based editing, AI speech generation for voice cloning, automatic filler word removal, and tools for screen recording and remote collaboration. The platform also supports multi-camera editing, transcription, and publishing features, enhancing the overall media creation experience. Descript serves a diverse clientele, including content creators, marketing teams, and enterprises, aiming to integrate video as a fundamental communication tool.

1902017

Genesys

Menlo Park, California, United States

Genesys, officially known as Genesys Cloud Services, Inc., is a prominent American software company that specializes in customer experience (CX) and contact center solutions. Founded in 1990, the company is headquartered in Menlo Park, California, and operates in over 50 countries. With a workforce of more than 6,000 employees, Genesys serves over 8,000 organizations globally, including Fortune 500 companies and government agencies across various sectors such as banking, healthcare, and retail. The company offers a range of omnichannel customer experience solutions through its cloud-native and hybrid platforms. These services enable businesses to engage customers via voice, chat, email, and social media while providing tools for workforce optimization, analytics, and automation. Key products include Genesys Cloud CX, a comprehensive cloud-native platform, and Genesys Multicloud CX, which supports both cloud and on-premises capabilities. Genesys is committed to integrating AI and machine learning into its offerings, enhancing customer interactions and operational efficiency.

82001990

Gladia

New York

Gladia is a Paris-based AI company founded in 2022 that specializes in audio intelligence. Its main product is the Audio Intelligence API, which allows developers and enterprises to integrate high-accuracy, real-time speech-to-text transcription, translation, and audio analytics into their applications. The API supports over 100 languages and offers features like speaker diarization, sentiment analysis, and emotion detection. The company aims to make audio intelligence accessible for any organization, regardless of industry or technology stack. Gladia serves a diverse range of clients, including over 600 enterprises and 70,000 users, with applications in real-time meeting assistance, contact centers, media production, and e-learning. The company operates on a pay-as-you-go pricing model, catering to both small developers and large enterprises. With a focus on a developer-first approach, Gladia ensures easy integration and reliable performance.

592022

Gridspace

Los Angeles, California

Gridspace is a technology company based in Los Angeles that specializes in contact center automation. The company develops a trusted platform that integrates voice networking and speech machine learning to enhance real-world conversations. Their mission focuses on creating agentic voice systems tailored for regulated industries, utilizing advanced techniques in speech recognition, natural language processing, and artificial intelligence. Founded in 2012 through a collaboration with SRI Speech Labs, Gridspace has pioneered innovative solutions for live conversation search and real-time analysis. Their primary services include automating voice calls for contact centers, providing executive tools for call analysis, and developing secure voice agents for enterprises, particularly in the financial services and insurance sectors. One of their key products, Gridspace Grace, is a conversational voice bot designed to manage customer interactions effectively. The company is recognized for its contributions to the field and serves some of the largest businesses globally, including trading desks and broadcasters.

122012

Hume AI

New York, New York

Hume AI is a research lab and technology company focused on developing empathic artificial intelligence that understands and responds to human emotions. Founded in March 2021 by Alan Cowen in New York, the company aims to integrate empathy into AI systems to enhance human emotional well-being. Hume AI creates multimodal AI models that interpret and generate emotionally nuanced speech. Its core products include the Empathic Voice Interface (EVI), which enables real-time voice interactions, and Octave, a text-to-speech system that delivers natural-sounding speech with emotional tone. The Expression Measurement API measures human expressive behavior across various media, documenting over 30 distinct emotions. Hume AI also offers an AI toolkit and a science-backed API for developers to build applications that better understand user intent and preferences. The company serves a diverse range of clients, including developers, enterprise software companies, healthcare providers, and customer service platforms, focusing on creating emotionally intelligent voice and expression-based experiences.

552021

Kits AI

San Francisco, California

Kits.AI, also known as Arpeggi Labs Inc., is an AI audio research company based in San Francisco. Founded in November 2021, the company launched its browser-based platform in July 2023, offering innovative music production tools. Kits.AI specializes in AI voice cloning, AI singing generation, and modernizing vocal workflows for music creators. The platform provides a range of features, including voice cloning, AI-generated singing vocals, and vocal isolation tools. It allows users to create custom voice models, correct pitch, and generate harmonies, all without the need for expensive hardware. Kits.AI serves a diverse audience, including music producers, vocalists, and content creators, with a focus on ethical AI practices by licensing artist voice models. The company has processed over 80 million minutes of vocals and has a user base of more than 7 million globally.

122021

Krisp

Berkeley, California

Krisp Technologies Inc. is a deep tech company founded in 2017 in Yerevan, Armenia, and now based in Berkeley, California. The company specializes in real-time Voice AI software that enhances digital voice communication by removing background noise, converting accents, and translating speech. Krisp's technology has processed over 4 trillion minutes of voice conversations and is used on more than 200 million devices globally. Krisp offers a comprehensive Voice AI platform featuring real-time noise cancellation, accent conversion, and voice translation in over 60 languages. Additional capabilities include automatic meeting transcriptions, AI meeting notes, audio enhancement, and tools for contact centers. The company also provides a developer SDK for integrating its voice models into third-party applications. Krisp serves a diverse clientele, including global call centers, telecommunications firms, business professionals, and developers, helping to improve communication quality and customer satisfaction.

2302017

Lalal.ai

Zug, Canton of Zug

LALAL.AI is a music technology company that specializes in AI-powered audio source separation and voice cleaning. Founded in 2020 and operated by Omnisale GmbH, the company has quickly grown to serve over 6.79 million registered users worldwide. LALAL.AI aims to simplify audio processing tasks, making high-quality results accessible to both professionals and creatives. The platform offers a suite of AI-driven tools that allow users to extract up to 10 different audio stems from any audio or video file without quality loss. Key features include AI stem separation, vocal removal, voice cleaning, and multi-stem support. LALAL.AI is available on various platforms, including web, desktop (Windows/Mac), mobile (iOS/Android), and through an API for integration into other applications. The company targets a diverse audience, including musicians, sound producers, podcasters, video editors, and content creators, providing essential tools for enhancing audio quality and clarity.

302020

Listnr

Middletown, Delaware

Listnr AI is a software development company founded in 2020, specializing in AI voice generation and text-to-video content creation. The platform offers a library of over 1,000 voices in 142 languages, enabling users to produce realistic audio and video quickly. With a user base of 3.5 million across more than 200 countries, Listnr AI serves creators, marketers, and businesses looking to enhance their content production. The company provides a comprehensive SaaS platform that includes features such as text-to-speech, voice cloning, text-to-video generation, AI podcast hosting, and AI dubbing. It also offers a TTS API for developers to integrate voice generation capabilities into their applications. Listnr AI operates on a tiered subscription model, catering to individual users, solo entrepreneurs, and agencies. The company is headquartered in Middletown, Delaware, with an operational office in Gurugram, India.

72020

LivePerson

New York, New York, United States

LivePerson is a global technology company based in New York City, specializing in conversational commerce and artificial intelligence (AI) software. Founded in 1995, the company is recognized for its flagship product, the Conversational Cloud, which enables direct messaging between consumers and brands, enhancing customer engagement and streamlining communication. The LivePerson Platform supports businesses in connecting with customers through various channels, including chat, voice, and social media. Key offerings include conversational AI tools, generative AI solutions, and data analytics for performance insights. LivePerson serves a diverse range of industries, such as financial services, retail, healthcare, travel, telecommunications, and automotive, helping brands improve customer interactions and operational efficiency. The company partners with major organizations like Avaya and Salesforce to deliver integrated solutions worldwide.

6101995

Mobvoi

Beijing, Beijing Shi

Mobvoi Information Technology Company Limited, founded in 2012 by former Google experts Zhifei Li and Yuanyuan Li, is an AI-first technology company based in Beijing, China. The company specializes in generative AI, voice interaction, and consumer electronics, aiming to enhance human-machine interaction. Mobvoi is publicly listed on the Hong Kong Stock Exchange and operates globally, distributing products in over 150 countries. Mobvoi's core offerings include advanced AI software solutions such as speech recognition, natural language understanding, and AI-generated content tools. Their product lineup features the TicWatch smartwatches, AI voice recorders like the TicNote, smart speakers, and wireless earbuds. The company serves both individual consumers and enterprise clients across various industries, providing customized AI solutions and digital employee generation.

2402012

Modulate

Somerville, Massachusetts

Modulate is a voice AI company founded in 2017 by MIT alumni, focused on creating prosocial voice intelligence technology. The company aims to enhance the safety and trust of online communications by addressing issues like toxicity and fraud. Modulate is recognized as a leader in conversational voice intelligence, particularly in dynamic and emotive real-world conversations. Its flagship products include Velma, an advanced AI model, and ToxMod, a proactive voice moderation platform. The company offers an audio-native AI platform that analyzes voice conversations, combining spoken words with acoustic signals to detect various risks, including fraud and deepfakes. Modulate serves a diverse range of enterprise customers, including video game studios like Activision and Riot Games, as well as call centers and online platforms. With its technology, Modulate has protected tens of millions of users and provided insights into hundreds of millions of conversations, improving user experience and safety across multiple industries.

522019

Nuance Communications

Burlington, Massachusetts

Nuance Communications is a leader in conversational artificial intelligence and ambient intelligence, based in Burlington, Massachusetts. As a Microsoft company, Nuance focuses on voice recognition solutions and natural language understanding technologies that enhance everyday work and life. The company offers a comprehensive range of conversational AI solutions, including clinical documentation tools like Dragon Medical One for healthcare providers. Its product lineup features Dragon NaturallySpeaking for speech recognition, Dragon Anywhere Mobile for mobile use, and specialized software like PowerScribe Mobile Radiologist for radiologists. Nuance also develops various AI-driven tools that integrate intelligence into daily workflows. Nuance serves a wide array of clients, including 77% of U.S. hospitals and 85% of Fortune 100 companies, demonstrating its strong presence in the healthcare and corporate sectors. Its technologies support professionals across healthcare, legal, business, and consumer domains, improving efficiency through advanced speech and language AI.

34001992

Otter.ai

Mountain View, California, United States

Otter.ai, Inc. is an American company based in Mountain View, California, specializing in AI-powered transcription and collaboration tools. Founded in February 2016, Otter.ai develops applications that convert voice data into searchable and actionable knowledge for both businesses and individuals. The company has processed over 1 billion meetings for more than 35 million users globally, generating significant customer ROI. The flagship software, Otter, offers real-time transcription with high accuracy, speaker recognition, and a searchable knowledge base. Key features include meeting summaries, action item generation, and integration with Zoom for seamless virtual meeting workflows. Otter.ai also provides a suite of products, including Otter Voice Notes, OtterPilot, and Otter AI Chat, catering to various customer segments such as business professionals, educators, journalists, and individuals. The company operates on a subscription-based model with tiered plans to meet diverse user needs.

3002016