FEAT

What FEAT does

FEAT (株式会社フィート, FEAT Limited) is a Tokyo-based audio and acoustic AI technology company founded in April 2005 as a spin-off from ATR (Advanced Telecommunications Research Institute International). Originally incorporated as CorpraTech and renamed to FEAT in 2010, the firm is a wholly-owned subsidiary of Silk Laboratory (株式会社シルク・ラボラトリ) following an October 2021 ownership restructuring. The company operates with a small core team (a hiring plan permitting up to three regular full-time employees) plus contracted R&D and business development support, with registered capital of ¥12,000,000 and a single Tokyo headquarters relocated in April 2025 to Shibadaimon, Minato-ku.

The company's technology platform, branded Sound Space Smart, integrates six core AI/audio capabilities: multilingual speech translation (30+ languages, NICT-licensed, with neural voice synthesis), area sound collection via X-Beamforming microphone arrays (Patent 7158976 with OKI Electric), Hidden Markov Model-based anomaly detection (Patents 6882397, 7377387 with Nissan and Toyo Technica), natural language processing with custom dictionary registration, multimodal emotion recognition, and evidence-based functional sound design. Products include consumer hearing-impaired communication apps (KoeTra with 1M+ downloads; SpeechCanvas with 393K+ downloads), enterprise editions (SpeechCanvas for Biz at ¥30,000/month), standalone hardware (Sound Pipette microphone at ¥600,000 per 4-set unit; PC Simultaneous Interpretation System Dokodemo Hon'yaku Neo at ¥15,600/month or ¥562,000 perpetual), and professional services (R&D contracting, generative AI integration consulting, and Sound Design services). Eleven patents and 10+ registered trademarks protect the portfolio.

FEAT monetizes through a mix of B2B enterprise subscriptions and licenses (SpeechCanvas for Biz, PC simultaneous interpretation), hardware unit sales (Sound Pipette), contracted R&D services for government agencies and universities, and joint development with major Japanese corporates (Nissan, OKI Electric, Toyo Technica, TOPPAN, KDDI). The free consumer apps (KoeTra, SpeechCanvas) are indirectly sponsored by the Ministry of Internal Affairs and Communications and six telecom operators, functioning as accessibility public goods and enterprise lead-generation channels. The customer base spans over 30 named enterprise and government organizations, including AEON Bank (146-branch SpeechCanvas deployment), Narita International Airport (NariTra since 2011), JR West (QuaVoice announcements), and Nissan Motor (anomaly detection co-development). The company holds ISO/IEC 27001:2022 certification and has received multiple Ministry of Internal Affairs and Communications recognitions for information accessibility.

Name

FEAT

Legal name

株式会社フィート

Website

https://feat-ltd.jp

Company type

Private

Founded year

2005

Operating status

Operating

Headcount range

1–10 employees

Short description

FEAT is a Tokyo-based audio and acoustic AI company offering multilingual speech translation, area sound collection microphones, AI anomaly detection, and hearing-impaired communication tools to Japanese enterprises and government clients.

Ownership category

akta.pro rank

Product category

Speech Translation & Acoustic AI Software

NAICS

Audio and Video Equipment Manufacturing (33431), Other Sound Recording Industries (512290), Sound Recording Studios (512240), Other Sound Recording Industries (51229)

SIC

Communications Equipment, Nec (3669), Telephone & Telegraph Apparatus (3661), Computer Communications Equipment (3576)

akta.pro primary industry

Audio & Speech Analytics (call analytics, QA, insights) (HDAAAFAH)

akta.pro secondary industries

On-Device Speech & Audio AI (wake word, ASR, enhancement) (HDAAAJAF), Voice Cloning, Dubbing & Audio Generative Media (HDAAAFAK), Classroom Audio, Voice Amplification & Communication Hardware/Software (EDAFAGAI), Assistive Listening & Classroom Audio (FM/DM, Soundfield) Systems (EDAKAIAE), Hearing Enhancement / Assistive Listening Audio Devices (CRACANAL), Microphones, Headsets & Audio Input Devices (HDACAFAH)

Keywords

• Multilingual speech translation
• Acoustic AI technology
• Hearing accessibility software
• Area sound collection
• Industrial anomaly detection

Headquarters

HQ city

Okubo

HQ country

Japan

HQ region

Asia

Offices1 record

Markets served

GTM type

B2B

Offering type

Software

Cost components

Technology or R&D, Personnel, Operations, Marketing or Sales

Revenue model

1. B2B Software Licensing and Sales (法人向け製品販売): Sales of enterprise software including SpeechCanvas for Biz (monthly subscription), PC simultaneous interpretation systems (monthly license or perpetual purchase), and area sound collection microphones (hardware). Revenue from system development and customization services.
2. Research and Development Services (研究開発): Contracted R&D services for government agencies (e.g., Ministry of Internal Affairs and Communications grants), universities, and corporations in acoustic analysis, multilingual translation systems, and sound design applications.
3. App-based Services (アプリケーションサービス): Free consumer apps (KoeTra, SpeechCanvas) with no direct revenue. Revenue indirectly generated through telecom operator sponsorships, government subsidies, and enterprise adoption of related paid services.
4. Hardware Sales (ハードウェア販売): Sale of Sound Pipette area sound collection microphones and related audio hardware. Minimum sales unit of 4 sets (60万 yen per set).

Pricing tiers

Go-to-market motion3 records

Distribution channels3 records

Marketing channels5 records

Core offering

FEAT develops and sells multilingual speech translation systems, hearing-impaired communication support applications, area sound collection microphones, and AI-based anomaly detection solutions built on proprietary audio and acoustic AI technologies. The company's offerings are centered on the "Sound Space Smart®" platform combining speech recognition, machine translation, speech synthesis, area sound collection, and acoustic anomaly detection for government, financial, manufacturing, transportation, and accessibility applications.

Product overview

FEAT (株式会社フィート) is a multilingual speech translation and acoustic engineering technology company offering a portfolio of products and services centered on 'Sound Space Smart®'. Core offerings include multilingual speech translation systems (多言語音声翻訳システム) and apps (VoiceBiz, どこでも翻訳®, NariTra), hearing-impaired communication support apps (こえとら, SpeechCanvas, SpeechCanvas for Biz), area sound collection devices (Sound Pipette® マイク), AI anomaly detection solutions (Sound Pipette® ターゲット), and functional sound design services. The company provides system development, application development, research and development, and server operations as services. All offerings are built on core technologies in speech recognition, machine translation, speech synthesis, acoustic processing, area sound collection, natural language processing, anomaly detection, and generative AI. The company holds ISO/IEC 27001:2022 certification and maintains partnerships with NICT, OKI Electric Industry, universities, and major Japanese corporations.

Differentiator

Problem solved

Functional benefit

Products and services

• SpeechCanvas for Biz (法人版) Paid business-grade hearing-impaired communication support application for government agencies, banks, enterprises, and schools. Features enhanced security (no server data storage), custom dictionary registration for proper nouns, and ID-based terminal management.
• PC Simultaneous Interpretation System (どこでも翻訳® Neo) Standalone simultaneous interpretation system combining multilingual interpretation server and client in a lightweight notebook PC. Supports 4-15 languages simultaneously depending on memory configuration (64GB: 4 languages, 128GB: 15 languages). Windows and Mac compatible, no internet required.
• Sound Pipette® Microphone (エリア収音マイク) Microphone array area sound collection hardware device using X-Beamforming® technology to extract audio from specific target areas while suppressing noise from non-target areas. Joint patent with OKI Electric Industry (Patent 7158976). Available in 6 colors, Windows/Linux software versions available.
• Sound Pipette® Target (AI異常検知ソリューション) AI anomaly detection solution combining Sound Pipette® area sound collection with Hidden Markov Model-based AI. Detects abnormal sounds probabilistically in equipment, automotive powertrains, and manufacturing lines. Co-developed with Nissan Motor and Toyo Technica (Patents 6882397, 7377387).
• KoeTra (こえとら) Free hearing-impaired communication support app converting speech to text and text to speech for hearing-impaired and hearing individuals. Supports iOS, Android, iPad. Downloaded over 1.2 million times by August 2025.
• SpeechCanvas (スピーチキャンバス) Face-to-face communication support app for hearing-impaired individuals at service counters, combining voice recognition (spoken words appear as text with furigana) and screen drawing input. Selected for Ministry of Internal Affairs 'Information Accessibility Good Examples 2023'.
• Dokodemo Hon'yaku® (どこでも翻訳) Serverless multilingual speech translation app for travel conversations that works without internet connection. Initial release with Japanese-English, Japanese-Chinese, and Japanese-Korean variants. Registered trademark No. 5864827.
• Multilingual Speech Translation System (多言語音声翻訳システム) Enterprise multilingual speech translation system providing voice translation app development, server construction, and operational maintenance. Supports speech recognition, machine translation, and speech synthesis for 30+ languages. Powers multilingual translation deployments for manufacturing and other enterprise clients.
• Sound Design (Sound Space Smart® / サウンドデザイン) Functional sound design service creating sounds that scientifically influence human behavior, cognition, and emotions. Includes functional sound design, anomaly sound detection systems, brand sound design, elderly care sound monitoring, acoustic consulting, and spatial acoustic design.
• Generative AI Consulting & Development (生成AI導入・開発支援) Generative AI consulting and development services combining generative AI with FEAT's speech translation, anomaly detection, emotion recognition, and area sound collection technologies. Includes PoC design, implementation, and ongoing support for context-aware summarization, meeting minutes auto-generation, and real-time report generation.

Quantifiable outcome

• KoeTra app reached 100万 (1 million) cumulative downloads, with 2025 reaching 122.5万 by August 2025
• +3 more outcomes

Named customers7 records

Segments7 records

Ideal customer profiles5 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability7 records

Feature6 records

Partnerships

Eight partnerships are on record, tiered major and core.

• 学校法人中央大学 (Chuo University)majorStrategic or Co-development Partner · 1 September 2025Ongoing joint research with Chuo University's Toi Laboratory on elderly care support, functional sound design, and acoustic monitoring. Published multiple joint papers at Japan Acoustical Society and Japan Society of Mechanical Engineers. Research on functional sounds for elderly behavioral change won Student Excellent Presentation Award 2025.
• 沖電気工業株式会社 (OKI Electric Industry)coreStrategic or Co-development Partner · 21 April 2025Joint development and patent co-filing for area sound collection technology (X-Beamforming) using microphone arrays. FEAT and OKI jointly hold Patent No. 7158976 for purpose area sound collection device. OKI's long-cultivated microphone array technology is integrated with FEAT's software for Sound Pipette microphone products.
• 株式会社IC (IC)majorStrategic or Co-development Partner · 1 January 2025Joint development partner for next-generation elderly monitoring service (実証試験 in Yokosuka City) and functional sound design research. FEAT's associate company (関連会社) listed alongside Silk Laboratory as related companies.
• 日产自動車株式会社 (Nissan Motor)coreStrategic or Co-development Partner · 1 January 2022Joint development of Hidden Markov Model-based anomalous sound detection system for automotive powertrains. Published in Society of Automotive Engineers of Japan Technical Papers 2023 (Vol. 54, No. 2). Joint patent holders (Patents 6882397, 7377387) for anomalous sound detection technology.
• 株式会社東陽テクnica (Toyo Technica)coreStrategic or Co-development Partner · 1 January 2022Joint development partner with Nissan Motor and FEAT for anomalous sound detection AI systems. Co-holder of patents 6882397 and 7377387 for anomalous sound detection systems, devices, methods, and programs.
• TOPPAN株式会社 (TOPPAN)majorTechnology or Integration · 1 January 2017FEAT provides development and operational support for VoiceBiz multilingual translation app for government and corporate use. Built on NICT's speech translation research with 12-language speech recognition/synthesis and 30-language machine translation. Ongoing service provision and maintenance.
• 国立研究開発法人情報通信研究機構 (NICT)coreTechnology or Integration · 1 January 2015NICT developed and transferred speech translation, speech recognition, and speech synthesis technologies to FEAT. FEAT licensed these technologies to build multiple products: NariTra (2011), VoiceTra+ (2012), KoeTra (2015), and SpeechCanvas (2017). NICT is a founding member of the Global Communication Development Promotion Council alongside FEAT.
• 総務省 / 電気通信事業者6社 (Ministry of Internal Affairs and Communications / 6 Telecom Operators)coreOthers · 1 January 2015KoeTra app is sponsored by Ministry of Internal Affairs and Communications plus 6 telecom operators (NTT Docomo, KDDI, SoftBank, Rakuten Mobile, NTT East, NTT West) who fund free server-side operations. Additionally, Daiwa Kosan (大東建託) sponsors as building rental operator.协賛制度 has been ongoing since 2015, with 100万 downloads achieved.

Scale indicators7 records

Recent moves7 records

Expansion highlights6 records

Broad incumbents

• Shure Incorporated: Shure manufactures professional microphones and audio equipment globally, including beamforming arrays for conference and broadcast applications. Comparable to FEAT's Sound Pipette microphone array on hardware capability but Shure is a broad incumbent serving professional audio markets rather than Japanese accessibility verticals.
• iFlytek: China-based AI speech recognition and translation company with broad industry vertical coverage. Comparable to FEAT on multilingual speech translation and accessibility education tools, but at global-enterprise scale versus FEAT's Japan focus.
• Microsoft (Azure AI Speech / Translator): Microsoft provides cloud-based speech translation and conversational AI across 100+ languages. Comparable to FEAT on multilingual speech translation but operates at vastly larger scale as a broad incumbent rather than a niche accessibility specialist.
• Cerence AI: Cerence provides AI for automotive in-cabin experiences including speech recognition. Comparable to FEAT's joint work with Nissan on anomalous sound detection, but Cerence focuses on the in-vehicle assistant and conversational AI layer rather than sound-based quality control.
• Google (Cloud Translation & Speech-to-Text): Google's translation and speech AI services compete with FEAT's core multilingual translation stack. Differs in cloud-only delivery model versus FEAT's offline/standalone PC interpretation, but represents the dominant competitive benchmark for translation accuracy.

Others

• OKI Electric Industry: OKI Electric jointly holds Patent 7158976 with FEAT for area sound collection microphone technology. As a major Japanese electronics manufacturer with microphone array heritage, OKI is a strategic partner and an alternative manufacturer for the Sound Pipette product line.
• TOPPAN Holdings: TOPPAN is both a partner (VoiceBiz development) and a comparator in multilingual translation services for government reception. Operates as a much larger systems integrator adjacent to FEAT's multilingual business, representing both a partner and a potential competitive channel.
• NICT (National Institute of Information and Communications Technology): Japanese government research institute that transferred core speech translation technology to FEAT. NICT is both a strategic partner and a comparator for advanced multilingual AI research; while not a commercial competitor, its research outputs shape FEAT's product roadmap.

Regional players

• Ubitus (AI services): Japan/Taiwan-based AI services company offering Japanese-language AI solutions including translation and accessibility. Comparable to FEAT as an Asian regional AI vendor focused on Japanese-language enterprise and consumer use cases, though Ubitus is broader in cloud gaming and AI services.

Direct peers

• Sourcenext (Pocketalk): Japanese creator of Pocketalk, the leading consumer handheld translation device. Directly comparable to FEAT's Dokodemo Hon'yaku on multilingual translation for travelers and accessibility use cases in Japan, with overlapping retail and enterprise channels.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat6 records

Key risks7 records

Key highlights7 records

Customer concentration

• LinkedInlinkedin.com/company/feat-limited

Compliance1 record

Revenue estimate

Valuation estimate

Number of profiles

Profiles4 records

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about FEAT