Smart recording pens that once cost thousands of yuan have now descended to the hundred-yuan market segment amid fierce competition from internet giants and numerous startups.

A credit card-sized ultra-thin device that magnetically attaches to the back of a phone, activated by pressing for two seconds to start recording, with a mobile app that quickly generates text transcripts and meeting summaries after recording ends - this magnetic card recorder was launched by Plaud, a Shenzhen startup established less than four years ago. Due to its lightweight convenience and comprehensive functionality, it has become a new market favorite in recording hardware, sparking imitation from internet giants and Huaqiangbei companies.

Compared to other AI hardware sectors, the AI recording pen market has a scale of only tens of billions of yuan and is considered a niche market. However, as AI large models accelerate their deployment, the AI recording pen market is expected to welcome a new growth cycle. Beyond price competition, what else can companies do?

**Plaud Returns from Overseas to Domestic Market**

On September 22, Plaud officially announced its entry into the domestic market while releasing multiple AI recording devices including Plaud Note Pro and Plaud NotePin S.

Plaud is a homegrown Shenzhen company that took an alternative path at startup, selling in regions and countries outside mainland China for the past few years, choosing a "Made in China - Overseas Sales - Return to Domestic" route.

In retrospect, this was a relatively wise decision. Domestic competition was fierce at the time, with leading companies represented by iFLYTEK holding significant advantages. Data shows that AI recording pens supporting real-time transcription currently account for over 35% of the Chinese market share; voice transcription technology conversion rates have risen from 68% in 2021 to 87% in 2025.

Plaud founder Xu Gao is a serial entrepreneur and investor with over 10 years of investment experience in the technology industry and overseas markets. He stated that the company has shipped over 1 million units globally, and the company's annualized revenue will reach $250 million in 2025, achieving double-digit growth compared to last year.

Plaud's return to the domestic market may be driven by sensing new business opportunities. Since this year, with the popularization of open-source large models, the threshold for voice-to-text conversion has dramatically lowered, putting large companies, small and medium enterprises, and even Huaqiangbei white-label merchants on the same starting line, all launching various AI recording card products.

Consumer electronics industry experts believe that one reason Plaud Note became a hit product is its construction of a complete user experience loop, from hardware audio collection and intelligent transcription to deep content summarization, forming a one-stop solution that eliminates the need to switch between different products, greatly improving efficiency and convenience.

On September 24, the Plaud Note was priced at 999 yuan and Plaud Note Pro at 1,299 yuan on brand flagship stores on JD.com, Tmall, and Douyin. However, product sales on various platforms have not shown explosive growth, with even flagship products selling only around a hundred units.

Seeing Plaud's overseas success last year, the domestic market has accelerated its follow-up this year.

In June, MOBVOI (02438.HK) launched the AI recording card TicNote, priced from 999 yuan. While AI recording cards were not originally created by MOBVOI, TicNote still received significant market attention.

On August 25, internet giants began making moves with competing products. Alibaba's DingTalk launched DingTalk A1 at its tenth anniversary conference, with the youth version priced as low as 499 yuan, supported by DingTalk's ecosystem office advantages to form a business closed loop. Additionally, 360, iFLYTEK, and others have successively released similar products.

When the commercial demand and monetization path for AI recording cards have been validated, industry-wide follow-up becomes inevitable.

Industry insiders believe that the ecosystem of tech hardware products often involves startups defining categories, giants leveraging ecosystem advantages to promote scale, and Huaqiangbei using supply chain advantages to quickly attack the sinking market. The industry may present a differentiated pattern of "high-end products competing on technology, mid-range competing on ecosystem, and low-end competing on cost."

**Domestic Market Forms Multi-dimensional Competitive Landscape**

As AI large models accelerate deployment and voice transcription technology matures, the AI hardware tools market welcomes a new growth cycle. Plaud's return to the domestic market adds another variable to the already competitive intelligent transcription sector.

Currently, the domestic intelligent transcription device market has formed a multi-dimensional competitive landscape, with leading manufacturers, emerging brands, and ecosystem players each showcasing their strengths.

As a representative of established industry manufacturers, iFLYTEK has been deeply cultivating the industry for years, relying on technological accumulation and brand foundation, focusing on serving professional fields such as healthcare and legal. Through deep cooperation with industry clients, it strengthens scenario adaptation capabilities, with product layouts emphasizing professionalism and brand trust construction.

Backed by internet giants like ByteDance and Alibaba, ecosystem products such as Feishu Miaoji and DingTalk Shanji rely on office collaboration platforms, deeply integrating transcription functions with task management and team sharing scenarios, providing seamless user experiences for ecosystem users, though they also face certain usage thresholds due to dependence on specific office systems.

Among emerging brands, represented by Tingnao AI, the choice is to break into the mass market with high cost-performance strategies, attracting individual users and small teams through simplified functional modules and optimized pricing systems, achieving market breakthrough through differentiated paths.

Unlike first-tier companies with vast ecosystem systems, AI startups like Plaud and MOBVOI focus more on creating differentiated experiences through "hardware + AI" vertical integration, targeting subdivided scenarios such as cross-border meetings, multilingual interviews, and new media content production, attempting to find survival space in giant-dominated markets through more precise scenario definitions and technical paths.

From current competitive focus, industry competition mainly concentrates on three dimensions: technical level, competing on transcription accuracy, processing speed, and multi-scenario adaptation capabilities; experience level, mainly around operational convenience, data security, and ecosystem integration; cost side presents layered pricing strategies of professional high-end versus inclusive affordable.

Despite continuous enrichment of market competition dimensions and respective advantages of leading and emerging players, from an industry perspective, intelligent transcription device users' needs for "vertical scenario customized services" and "cross-device collaboration" have not been fully satisfied.

Domestic hardware manufacturer competition is extremely fierce, presenting Plaud with a series of challenges.

There's a consensus in the hardware industry: no company can continuously create hit products. Hardware replacement cycles are longer than fast-moving consumer goods, and user decisions are more weighted. Unless maintaining leapfrog upgrades, bottlenecks are quickly encountered.

Facing such market environment, Kong Chi, head of Plaud's global sales center, stated they "pay attention but won't overly worry about competition." He believes the industry is still in a development period, hoping everyone can collectively expand the AI product sector through product strength and concepts.

Zhang Xiaorong, director of the Deep Technology Research Institute, suggests that in fierce homogenization competition in the AI industry, startups must take the route of "precise positioning + differentiated technology + scenario deep cultivation" to survive. "Product lines should avoid trying to cover everything; technically, don't aim to be 'omnipotent AI.' Start by becoming synonymous with a specific subdivided field, make good products, strengthen experience, and reduce costs."