Pith. sign in

REVIEW 4 major objections 6 minor 61 references

Towards Energy-Efficient and Low-Latency Voice-Controlled Smart Homes: A Proposal for Offline Speech Recognition and IoT Integration

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Offline keyword-spotting chips in every appliance could end smart-home cloud lag

desk verdict A plainly written proposal for per-appliance offline KWS with a brokerless mesh; the architecture is coherent, but every headline benefit is unmeasured. read the letter →

arxiv 2506.07494 v2 pith:ZA2BUTDN submitted 2025-06-09 cs.SD cs.CYeess.AS

Peng Huang , Imdad Ullah , Xiaotong Wei , Tariq Ahamed Ahanger , Najm Hassan , Zawar Hussain Shah This is my paper
classification cs.SDcs.CYeess.AS
keywords offlinespeechrecognitionkeywordspottingsmarthomeIoTdecentralizednetworklowlatencyenergyefficiencysinglepointoffailure
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Cloud-based voice assistants make smart-home commands travel to remote servers and back, which costs energy, adds latency, and leaves the system helpless if the network or a central hub fails. This position paper proposes to put offline keyword spotting (KWS) directly into household appliances so each device can hear and understand voice commands locally. It combines that with a decentralized local IoT network, using mesh networking and the CoAP protocol, so devices cooperate without a central server. The paper argues that this design yields low-latency voice control anywhere in the home without Internet dependence, with better scalability and energy sustainability. The proposal rests on prior work demonstrating KWS on microcontrollers and on commercial chips such as the Voitist 811.

What carries the argument

The load-bearing mechanism is the keyword-spotting unit embedded in each appliance, paired with a decentralized local mesh network. The KWS unit replaces the need for a remote speech-recognition server by recognizing wake-up and command words on-device, and the mesh network lets a command heard by one device reach a target device elsewhere in the home. The paper identifies two hardware patterns (coexist and unified integration) and names CoAP as the application protocol that fits the decentralized design because it allows direct device-to-device messaging without a broker.

What would settle it

Measure the wake-word accuracy and false-trigger rate of a $1.50-class keyword-spotting chip at typical speaking distances of two to five meters in a home with background noise; if accuracy falls below what users expect from cloud assistants, or if the always-on listening draws more power than the appliance's idle budget, the central claim does not survive.

Watch

Extended reading notes

Core claim

The central claim is that the standard cloud round-trip for voice control is unnecessary, and that a home can instead give every appliance its own always-listening keyword-spotting unit. Two integration routes are described: the coexist approach, which adds a separate KWS microcontroller beside the appliance's existing controller over UART, I2C, or SPI, and the unified approach, which replaces the controller with a single AI chip that contains an NPU, codec, and storage, such as the Voitist 811 with about 300 keyword slots and 0.2-second feedback. The appliances then form a decentralized mesh network using BLE Mesh, ZigBee, or Wi-Fi Mesh at the lower layers and CoAP for application-layer messaging, so no broker or hub is a single point of failure. Three interaction modes are defined: direct device interaction, cross-device interaction within a room, and cross-device interaction between rooms. The paper's conclusion is that this architecture gives low-latency, energy-efficient, robust voice control that works without the Internet.

Load-bearing premise

The load-bearing premise is that a small, cheap keyword-spotting unit can be embedded in every appliance and will hear voice commands reliably from normal speaking positions while staying within each appliance's power and cost budget.

Editorial extensions

If this is right

  • Voice commands are processed at the device, so response time no longer depends on Internet speed or cloud server load.
  • If one device loses power or fails, only that device's functions are lost; other appliances continue to accept voice commands.
  • Adding or removing appliances does not require reconfiguring a central hub, because the mesh self-heals and new nodes join existing neighbors.
  • Away from reliable Internet, or after an outage, the smart home keeps its voice-control functions rather than going silent.
  • The per-device cost of voice control can be small enough to embed in ordinary appliances, eliminating the need for a separate smart speaker.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If every appliance listens continuously, aggregate standby power and false-trigger rates become the real design constraint; the paper does not measure either, but the proposal stands or falls on those numbers.
  • A natural next step is to benchmark a Voitist-811-class chip against a cloud assistant in a real room, measuring wake-word accuracy, false accepts, and latency at normal speaking distances and background noise.
  • Keeping voice data entirely local is an implicit privacy benefit the paper mentions only indirectly; it could make offline KWS attractive even where the Internet is reliable.
  • The 300-keyword limit per chip suggests a need for a shared command vocabulary or hierarchical wake words, otherwise larger homes will exhaust the keyword budget quickly.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This position paper proposes an offline (Internet-independent) voice-controlled smart home architecture. The proposal has two pillars: (1) integrating keyword-spotting (KWS) units directly into household appliances, via either a separate KWS MCU ('Coexist Integration Approach') or a single AI chip that also replaces the appliance controller ('Unified Integration Approach'), and (2) replacing the cloud-dependent, hub-based control path with a decentralized local IoT mesh network using technologies such as BLE Mesh, ZigBee, and Wi-Fi Mesh with CoAP for application-layer messaging. The paper describes a four-layer conceptual architecture, categorizes appliance components, defines three voice-interaction modes (direct, cross-device within a subsystem, and cross-device between subsystems), and discusses scalability and robustness. It contains no experiments, measurements, or executable system; the only quantitative hardware evidence is the Voitist 811 vendor datasheet, and Section V explicitly acknowledges the lack of large-scale practical testing. The central claims are that the proposal provides lower latency, reduced energy consumption, and improved robustness compared with cloud-based systems such as Amazon Alexa.

Significance. If the stated claims could be substantiated, the proposal would address a real gap: most offline smart-home voice systems reviewed in Section III.B use a centralized local hub, and the decentralized, KWS-per-appliance architecture is a plausible conceptual alternative that avoids a hub single point of failure. The manuscript is clearly organized, and the authors are honest in Section V about the absence of large-scale testing. I also credit the paper for explicitly discussing three interaction modes and for acknowledging (in Section IV.E) the distance limitation of direct interaction. However, the paper's contribution is architectural only; it does not provide a quantitative energy model, a latency budget, or an acoustic coverage analysis. Because the abstract states that the proposal 'will' deliver low-latency, energy-sustainable operation, the current evidence is insufficient to support those performance claims. The proposal is falsifiable in principle, and the missing analyses could be supplied in a revision.

major comments (4)
  1. [§IV.B, §V, Abstract] The central claim that the proposed system reduces energy consumption compared with cloud-based systems is not supported by any quantitative analysis. Section IV.B acknowledges that integrating a KWS unit 'may alter the DC power requirements' and must be 'evaluated and adapted,' but no such evaluation is provided. The cost model also omits the always-on power of the per-device microphone front end, codec, KWS MCU, and mesh radio (including idle listening), which scales with the number of appliances. Without a system-level standby-power budget comparing, say, a 20-device home against a baseline Echo in wake-word mode plus ZigBee peripherals, the abstract's 'energy sustainability' and Section V's 'reduces unnecessary energy consumption' are unsupported; a fully distributed always-listening network could plausibly exceed the baseline.
  2. [§IV.C, Table I] The evidence cited for low-power per-device KWS feasibility is incomplete. The 2.51 µW figure from reference [32] (Table I) is a 40nm ASIC simulation of the LSTM accelerator alone, not the full signal chain including microphone, ADC, codec, host MCU, and radio. The Voitist 811 datasheet [45] is cited for cost and keyword capacity, but its operating power is not reported in the paper. Using these sources as evidence that embedding KWS in every appliance is energy-feasible overstates what is known. The authors should report or estimate the end-to-end always-on power of a complete KWS node and relate it to the appliance's existing standby power.
  3. [§IV.E, §V, Algorithms 1-2] The latency claim is similarly unquantified. The only numeric figure in the proposal is the Voitist 811's 'feedback within 0.2 seconds' (Section IV.C), which is a component-level response, not an end-to-end user-perceived latency for direct, cross-device, or between-subsystem interactions; the latter involves wake-word detection, command recognition, network hops, and action execution, none of which are measured. Algorithms 1 and 2 are pseudocode sketches and cannot serve as a basis for a latency analysis. To support the 'low-latency voice control' claim, the authors should either report end-to-end measurements or characterize the expected latency budget with a concrete decomposition.
  4. [§IV.D] The robustness claim is weakened by the gateway role described in Section IV.D. The text states that gateways act as bridges between different IoT technologies (e.g., ZigBee and Wi-Fi); these gateways are special nodes whose failure would partition the network and prevent cross-technology communication, effectively reintroducing single points of failure at the protocol boundary. The paper should clarify how the decentralized architecture handles gateway failure, or restrict the robustness claim to individual homogeneous networks.
minor comments (6)
  1. [§IV.D] The citation placeholder '[ansari2018Internet]' appears in the text but is not in the reference list; replace it with a proper numbered reference or remove it.
  2. [Tables I and II] Both tables use the placeholder caption 'CAPTION'; provide descriptive captions that summarize the content of each table.
  3. [Reference [30]] Reference [30] is incomplete: it has no venue, year, or publisher listed. Please complete the bibliographic information.
  4. [Algorithms 1 and 2] In Algorithm 1, line 16 says 'Updatet until it reaches0' without specifying the update rule; clarify with a concrete expression such as 't ← t − 1 per loop iteration' or equivalent.
  5. [§III.B] The text says 'As discussed in Section 2' but the paper uses Roman numerals for sections; this should be 'Section II' for consistency.
  6. [Figure 5] The abbreviations 'SPK' and 'PA' in Figure 5 are not defined in the caption or in the main text; please define them in the figure caption.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is a design position paper whose energy and latency claims are unvalidated premises, not derived predictions.

full rationale

This paper proposes an architecture, it does not perform a derivation. There are no fitted parameters, no predicted quantities, and no equations whose outputs reduce to their inputs. The central claims (low-latency offline control, energy sustainability, scalability, robustness) are asserted as benefits of the proposed design and are supported, where anything is supported at all, by external citations: independent KWS literature, standard IoT protocol references, and a vendor datasheet for the Voitist 811 chip. No load-bearing step is justified by a self-citation chain or by a uniqueness theorem imported from the authors' own prior work. The paper is honest about its status: Section IV.B states that integrating a KWS unit 'may alter the DC power requirements. Therefore, it is necessary to evaluate and adapt the power components to meet these new requirements,' which concedes an open design issue rather than claiming a derived result. Section V explicitly says 'The limitations of this study include the lack of large-scale practical application testing.' These admissions confirm that the energy-sustainability and performance benefits are proposals awaiting measurement, not conclusions forced by construction. The skeptic's concern that always-on per-device KWS units and mesh radios could exceed a cloud hub's standby power is a legitimate empirical risk, but an unsupported premise is not a circularity. Accordingly, no circular steps are found and the score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The ledger is empty of free parameters and invented entities because the paper makes no quantitative measurements. It rests on unvalidated domain assumptions about embedded KWS accuracy, mesh reliability, brokerless CoAP messaging, and room-based command routing.

assumptions (5)
  • domain assumption Keyword spotting can run with acceptable accuracy on resource-constrained hardware embedded in household appliances.
    Section IV.C relies on prior small-footprint KWS work [29-33] and the Voitist 811 datasheet [45] to argue feasibility. No in-appliance acoustic evaluation is provided.
  • domain assumption Mesh network technologies (BLE Mesh, ZigBee, Wi-Fi Mesh) provide the robustness, scalability, and self-healing claimed for home deployment.
    Section IV.D cites standard protocol surveys [50,51], but does not test behavior under expected home node counts, traffic, or interference.
  • domain assumption CoAP over UDP can support decentralized device-to-device communication without a broker and with acceptable overhead.
    Section IV.D selects CoAP over MQTT based on published comparisons [57,59]; no implementation validates this choice in the proposed system.
  • domain assumption The room attribute filling and cross-device routing in Section IV.E uniquely determines the intended target device.
    The design assumes the listening device can infer the current room and route commands correctly, but does not specify how conflicting or overlapping commands from multiple KWS devices are resolved.
  • ad hoc to paper A single low-cost AI chip such as the Voitist 811 can perform the appliance controller role as well as KWS, with acceptable power and latency.
    Section IV.C bases the Unified Integration Approach on a vendor datasheet [45]; no independent benchmark or energy measurement is reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Energy-Efficient and Low-Latency Voice-Controlled Smart Homes: A Proposal for Offline Speech Recognition and IoT Integration." pith.science (2026). https://pith.science/paper/ZA2BUTDN

@misc{pith2026250607494,
  author       = {Pith},
  title        = {Pith review of: Towards Energy-Efficient and Low-Latency Voice-Controlled Smart Homes: A Proposal for Offline Speech Recognition and IoT Integration},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZA2BUTDN}},
  note         = {Machine review of arXiv:2506.07494}
}
read the original abstract

The smart home systems, based on AI speech recognition and IoT technology, enable people to control devices through verbal commands and make people's lives more efficient. However, existing AI speech recognition services are primarily deployed on cloud platforms on the Internet. When users issue a command, speech recognition devices like ``Amazon Echo'' will post a recording through numerous network nodes, reach multiple servers, and then receive responses through the Internet. This mechanism presents several issues, including unnecessary energy consumption, communication latency, and the risk of a single-point failure. In this position paper, we propose a smart home concept based on offline speech recognition and IoT technology: 1) integrating offline keyword spotting (KWS) technologies into household appliances with limited resource hardware to enable them to understand user voice commands; 2) designing a local IoT network with decentralized architecture to manage and connect various devices, enhancing the robustness and scalability of the system. This proposal of a smart home based on offline speech recognition and IoT technology will allow users to use low-latency voice control anywhere in the home without depending on the Internet and provide better scalability and energy sustainability.

Figures

Figures reproduced from arXiv: 2506.07494 by the authors.

Figure 1
Figure 1. A typical smart home system, where smart home devices connect to the Echo Speaker via local connectivity protocols such as BLE Mesh, ZigBee, [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. A typical smart home system, where smart home devices are connected to the manufacturer’s cloud. The user issues a request via the Echo Speaker, [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The architecture of the smart home system can be divided into four layers: System Layer, Subsystem Layer, Module Layer, and Device Layer. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Below is an explanation of each component. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 4
Figure 4. Figure 4: The internal elements of household appliances can be categorized into five components based on their purposes. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Two approaches to integrating KWS units into household appliances: The Coexist Integration Approach establishes a communication interface between [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: The subscription-publish model of the MQTT protocol involves [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 8
Figure 8. Figure 8: Direct device interaction, where the user interacts directly with the [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]
Figure 9
Figure 9. Figure 9: Cross-device Interaction (within a subsystem), where the user interacts [PITH_FULL_IMAGE:figures/full_fig_p010_9.png]
Figure 10
Figure 10. Figure 10: Cross-device Interaction (between subsystems), where the user [PITH_FULL_IMAGE:figures/full_fig_p010_10.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

61 extracted references · 56 canonical work pages

  1. [32]

    Owsnet: Towards real-time offensive words spotting network for consumer iot devices

    Bharath Sudharsan et al. “Owsnet: Towards real-time offensive words spotting network for consumer iot devices”. In:2021 IEEE 7th World Forum on Internet of Things (WF-IoT). IEEE. 2021, pp. 83–88

    work page 2021

  2. [45]

    1-Wire Communication Protocol for Debugging Modem Chipsets

    T Ninikrishna et al. “1-Wire Communication Protocol for Debugging Modem Chipsets”. In:Emerging Re- search in Computing, Information, Communication and Applications: ERCICA 2015, Volume 2. Springer. 2016, pp. 535–548

    work page 2015

  3. [1]

    Application of speech recognition technology in IoT smart home

    Peng Wang et al. “Application of speech recognition technology in IoT smart home”. In:2019 IEEE 3rd Ad- vanced Information Management, Communicates, Elec- tronic and Automation Control Conference (IMCEC). 2019, pp. 1264–1267.DOI: 10 . 1109 / IMCEC46724 . 2019.8984175

    work page arXiv 2019

  4. [2]

    Evaluating the Efficiency of V oice Control as Human Machine Interface in Production

    Marvin Norda et al. “Evaluating the Efficiency of V oice Control as Human Machine Interface in Production”. In:IEEE Transactions on Automation Science and En- gineering(2023), pp. 1–12.DOI: 10.1109/TASE.2023. 3302951

    doi:10.1109/tase.2023 2023

  5. [3]

    Design and evaluation of a smart home voice interface for the elderly: acceptability and objection aspects

    Franc ¸ois Portet et al. “Design and evaluation of a smart home voice interface for the elderly: acceptability and objection aspects”. In:Personal and Ubiquitous Computing17 (2013), pp. 127–144

    work page 2013

  6. [4]

    Tajammul Pangarkar.Smart Speaker Statistics: Best Voice Control Technology. Web. Accessed: 2024-06-20. June 2024.URL: https://scoop.market.us/smart-speaker- statistics/

    work page 2024

  7. [5]

    The Insecurity of Home Digital V oice Assistants - Vulnerabilities, Attacks and Countermea- sures

    Xinyu Lei et al. “The Insecurity of Home Digital V oice Assistants - Vulnerabilities, Attacks and Countermea- sures”. In:2018 IEEE Conference on Communications and Network Security (CNS). 2018, pp. 1–9.DOI: 10. 1109/CNS.2018.8433167

    work page arXiv 2018

  8. [6]

    Amazon.Amazon Alexa Voice AI — Alexa Developer Official Site. Web. Accessed: 2024-06-23. June 2024. URL: https://developer.amazon.com/en-US/alexa

    work page 2024

Show all 61 references
  1. [7]

    Amazon.Alexa Built-in Devices Official Site: Create Voice-First Devices. Web. Accessed: 2024-06-23. June 2024.URL: https://developer.amazon.com/en-US/alexa/ devices/alexa-built-in

    2024

  2. [8]

    Amazon.AVS UX Attention System —Alexa Voice Ser- vice. Web. Accessed: 2024-06-23. June 2024.URL: https://developer.amazon.com/en-US/docs/alexa/alexa- voice-service/ux-design-attention.html

    2024

  3. [9]

    Making Your Devices Speak. Inte- gration between Amazon Alexa and the Managed IoT Cloud

    Thomas Holden. “Making Your Devices Speak. Inte- gration between Amazon Alexa and the Managed IoT Cloud”. MA thesis. UiT Norges arktiske universitet, 2018

    2018

  4. [10]

    A Review on Client-Server based applications and research opportunity

    Santosh Kumar. “A Review on Client-Server based applications and research opportunity”. In:International Journal of Recent Scientific Research10.7 (2019), pp. 33857–3386

    2019

  5. [11]

    Real time video streaming from multi-source using client-server for video distribution

    Ibrahim MI Zebari, Subhi RM Zeebaree, and Ha- jar Maseeh Yasin. “Real time video streaming from multi-source using client-server for video distribution”. In:2019 4th Scientific International Conference Najaf (SICN). IEEE. 2019, pp. 109–114

    2019

  6. [12]

    Understanding User Perceptions of Robot’s Delay, V oice Quality-Speed Trade-off and GUI during Conversation

    Zhenhui Peng et al. “Understanding User Perceptions of Robot’s Delay, V oice Quality-Speed Trade-off and GUI during Conversation”. In:Extended Abstracts of the 2020 CHI Conference on Human Factors in Computing Systems. 2020, pp. 1–8

    2020

  7. [13]

    Carsen Holaday.Amazon Alexa down updates — Users complain voice assistant connectivity and control are not working as outage underway — The US Sun. Web. Accessed: 2024-06-25. June 2024.URL: https://www. the - sun . com / tech / 11807249 / amazon - alexa - down - updates-not-working/

    2024

  8. [14]

    Sabrina Johnson.Amazon’s Alexa ’down for thousands of customers’ — Metro News. Web. Accessed: 2024- 06-25. Apr. 2022.URL: https://metro.co.uk/2022/04/ 16/amazons-alexa-down-for-thousands-of-customers- 16479166/

    2024

  9. [15]

    Smart home technologies in Europe: A critical review of concepts, benefits, risks and policies

    Benjamin K Sovacool and Dylan D Furszyfer Del Rio. “Smart home technologies in Europe: A critical review of concepts, benefits, risks and policies”. In:Renewable and sustainable energy reviews120 (2020), p. 109663

    2020

  10. [16]

    Amazon.Local Connectivity Protocol Options for Smart Home Devices — Alexa Skills Kit. Web. Accessed: 2024-06-23. June 2024.URL: https://developer.amazon. com/en- US/docs/alexa/smarthome/wwa- connection- options.html

    2024

  11. [17]

    The Bluetooth mesh standard: An overview and experimental evaluation

    Mathias Baert et al. “The Bluetooth mesh standard: An overview and experimental evaluation”. In:Sensors18.8 (2018), p. 2409

    2018

  12. [18]

    Connected smart home over matter protocol

    Wondimu Zegeye, Ahamed Jemal, and Kevin Kornegay. “Connected smart home over matter protocol”. In:2023 IEEE International Conference on Consumer Electron- ics (ICCE). IEEE. 2023, pp. 1–7

    2023

  13. [19]

    ZigBee wire- less standard

    Stanislav Safaric and Kresimir Malaric. “ZigBee wire- less standard”. In:Proceedings ELMAR 2006. IEEE. 2006, pp. 259–262

    2006

  14. [20]

    Provisioning technical interoperability within ZigBee and BLE in IoT environment

    Taibur Rahman and Swarnendu Kumar Chakraborty. “Provisioning technical interoperability within ZigBee and BLE in IoT environment”. In:2018 2nd Inter- national Conference on Electronics, Materials Engi- neering & Nano-Technology (IEMENTech). IEEE. 2018, pp. 1–4

    2018

  15. [21]

    Amazon.Amazon.com.au: Echo & Alexa devices: Ama- zon Devices & Accessories. Web. Accessed: 2024-06-

    2024

  16. [22]

    Amazon.Understand Smart Home Skills — Alexa Skills Kit. Web. Accessed: 2024-06-25. June 2024.URL: https: //developer.amazon.com/en-US/docs/alexa/smarthome/ understand-the-smart-home-skill-api.html

    2024

  17. [23]

    Amazon.Smart Home Development Options — Alexa Skills Kit. Web. Accessed: 2024-06-25. June 2024.URL: https : / / developer. amazon . com / en - US / docs / alexa / smarthome/development-options.html

    2024

  18. [24]

    Deep spoken keyword spot- ting: An overview

    Iv ´an L ´opez-Espejo et al. “Deep spoken keyword spot- ting: An overview”. In:IEEE Access10 (2021), pp. 4169–4199

    2021

  19. [25]

    June 2024.URL: https : / / www. amazon . com . au / Alexa- Amazon- Echo- Smart- Speakers/b?ie=UTF8& node=5425434051

    2024

  20. [26]

    Continuous hidden Markov modeling for speaker-independent word spotting

    Jan Robin Rohlicek et al. “Continuous hidden Markov modeling for speaker-independent word spotting”. In: International Conference on Acoustics, Speech, and Signal Processing,IEEE. 1989, pp. 627–630

    1989

  21. [27]

    Small-footprint keyword spotting using deep neural networks

    Guoguo Chen, Carolina Parada, and Georg Heigold. “Small-footprint keyword spotting using deep neural networks”. In:2014 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE. 2014, pp. 4087–4091. 13

    2014

  22. [28]

    Convolutional neural networks for small-footprint keyword spotting

    Tara N Sainath and Carolina Parada. “Convolutional neural networks for small-footprint keyword spotting.” In:Interspeech. 2015, pp. 1478–1482

    2015

  23. [29]

    Low-latency convolutional recurrent neu- ral network for keyword spotting

    Hu Du et al. “Low-latency convolutional recurrent neu- ral network for keyword spotting”. In:2018 Joint 10th International Conference on Soft Computing and Intelli- gent Systems (SCIS) and 19th International Symposium on Advanced Intelligent Systems (ISIS). IEEE. 2018, pp. 802–807

    2018

  24. [30]

    Hello edge: Keyword spotting on microcontrollers

    Yundong Zhang et al. “Hello edge: Keyword spotting on microcontrollers”. In:arXiv preprint arXiv:1711.07128 (2017)

    2017 arXiv

  25. [31]

    V oice Recognition on Simple Microcontrollers

    Callie Y Kim, Diana Chou, and Geng Liu. “V oice Recognition on Simple Microcontrollers”. In: ()

  26. [33]

    A 2.5µW KWS engine with pruned LSTM and embedded MFCC for IoT applica- tions

    Yi Sheng Chong et al. “A 2.5µW KWS engine with pruned LSTM and embedded MFCC for IoT applica- tions”. In:IEEE Transactions on Circuits and Systems II: Express Briefs69.3 (2021), pp. 1662–1666

    2021

  27. [34]

    Sub-mW keyword spotting on an MCU: Analog binary feature extraction and binary neural networks

    Gianmarco Cerutti et al. “Sub-mW keyword spotting on an MCU: Analog binary feature extraction and binary neural networks”. In:IEEE Transactions on Circuits and Systems I: Regular Papers69.5 (2022), pp. 2002– 2012

    2022

  28. [35]

    Offline voice detection in smart homes

    Sina Ojaghi et al. “Offline voice detection in smart homes”. In:The 4th International Conference on Elec- trical Engineering, Computer, Mechanics and Artificial Intelligence. 2023

    2023

  29. [36]

    Application of speech recognition technology in IoT smart home

    Peng Wang et al. “Application of speech recognition technology in IoT smart home”. In:2019 IEEE 3rd Ad- vanced Information Management, Communicates, Elec- tronic and Automation Control Conference (IMCEC). IEEE. 2019, pp. 1264–1267

    2019

  30. [37]

    Electrical household appliances con- trol using voice command based on microcontroller

    Nefy Puteri Novani, Mohammad Hafiz Hersyah, and Ryon Hamdanu. “Electrical household appliances con- trol using voice command based on microcontroller”. In:2020 International Conference on Information Tech- nology Systems and Innovation (ICITSI). IEEE. 2020, pp. 288–293

    2020

  31. [38]

    Next generation home automation system based on voice recognition

    ME Moeid, HS Hana, and RZ Amer. “Next generation home automation system based on voice recognition”. In:ICEMIS20 (), pp. 1–7

  32. [39]

    HomeIO: Offline smart home automation system with automatic speech recognition and household power usage tracking

    IB Chandra Irugalbandara et al. “HomeIO: Offline smart home automation system with automatic speech recognition and household power usage tracking”. In: 2022 IEEE World AI IoT Congress (AIIoT). IEEE. 2022, pp. 571–577

    2022

  33. [40]

    Accessed: 2024-07-05

    NXP Semiconductors.I2S bus specification. Accessed: 2024-07-05. 2022.URL: https://www.nxp.com/docs/en/ user-manual/UM11732.pdf

    2024

  34. [41]

    Accessed: 2024-07-05

    NXP Semiconductors.I2C-bus specification and user manual. Accessed: 2024-07-05. 2021.URL: https : / / www.nxp.com/docs/en/user-guide/UM10204.pdf

    2024

  35. [42]

    Ac- cessed: 2024-07-05

    Analog Devices.Introduction to SPI Interface. Ac- cessed: 2024-07-05. 2018.URL: https : / / www . allelcoelec.in/datasheets.b8/ADG1412YCPZ- REEL7. pdf

    2024

  36. [43]

    Uni- versal asynchronous receiver and transmitter (uart)

    Umakanta Nanda and Sushant Kumar Pattnaik. “Uni- versal asynchronous receiver and transmitter (uart)”. In: 2016 3rd international conference on advanced com- puting and communication systems (ICACCS). V ol. 1. IEEE. 2016, pp. 1–5

    2016

  37. [44]

    Wired-FM, a novel distributed digital single-wire field bus

    Hamid Fahim Rezaei and Anton Kruger. “Wired-FM, a novel distributed digital single-wire field bus”. In:2012 IEEE International Conference on Electro/Information Technology. 2012, pp. 1–4.DOI: 10 . 1109 / EIT. 2012 . 6220723

    2012

  38. [46]

    Accessed: 2024- 07-08

    intenginetech.VOI811 Datasheet V2.3. Accessed: 2024- 07-08. 2023.URL: https://www.intenginetech.com/

    2024

  39. [47]

    An Approach of BLE Mesh Network For Smart Home Application

    Quoc Thi Tran et al. “An Approach of BLE Mesh Network For Smart Home Application”. In:2020 In- ternational Conference on Advanced Computing and Applications (ACOMP). 2020, pp. 170–174.DOI: 10. 1109/ACOMP50827.2020.00034

    2020

  40. [48]

    Design of Smart Home System Based on ZigBee

    Su Zhihua. “Design of Smart Home System Based on ZigBee”. In:2016 International Conference on Robots & Intelligent System (ICRIS). 2016, pp. 167–170.DOI: 10.1109/ICRIS.2016.35

    2016 doi

  41. [49]

    Evaluation of Full-Mesh Networks for Smart Home Applications

    V o Que Son and Nguyen Tho Anh Khoa. “Evaluation of Full-Mesh Networks for Smart Home Applications”. In:2019 International Symposium on Electrical and Electronics Engineering (ISEE). 2019, pp. 73–78.DOI: 10.1109/ISEE2.2019.8920920

    2019

  42. [50]

    Development of WiFi Mesh Infrastructure for Internet of Things Applications

    Rifki Muhendra et al. “Development of WiFi Mesh Infrastructure for Internet of Things Applications”. In: Procedia Engineering170 (2017). Engineering Physics International Conference 2016 – EPIC 2016, pp. 332– 337.ISSN: 1877-7058.DOI: https://doi.org/10.1016/j. proeng.2017.03.0...

    2017 doi

  43. [51]

    Bluetooth Mesh Anal- ysis, Issues, and Challenges

    ´angela Hern ´andez-Solana et al. “Bluetooth Mesh Anal- ysis, Issues, and Challenges”. In:IEEE Access8 (2020), pp. 53784–53800.DOI: 10 . 1109 / ACCESS . 2020 . 2980795

    2020

  44. [52]

    Bluetooth low energy mesh networks: A survey

    Seyed Mahdi Darroudi and Carles Gomez. “Bluetooth low energy mesh networks: A survey”. In:Sensors17.7 (2017), p. 1467

    2017

  45. [53]

    Study on ZigBee network architecture and routing algorithm

    Jianpo Li et al. “Study on ZigBee network architecture and routing algorithm”. In:2010 2nd International Conference on Signal Processing Systems. V ol. 2. 2010, pp. V2-389-V2–393.DOI: 10 . 1109 / ICSPS . 2010 . 5555486

    2010

  46. [54]

    A Comparative Study of Wireless Protocols: Bluetooth, UWB, ZigBee, and Wi-Fi

    Jin-Shyan Lee, Yu-Wei Su, and Chung-Chou Shen. “A Comparative Study of Wireless Protocols: Bluetooth, UWB, ZigBee, and Wi-Fi”. In:IECON 2007 - 33rd Annual Conference of the IEEE Industrial Electronics Society. 2007, pp. 46–51.DOI: 10.1109/IECON.2007. 4460126. 14

    2007 doi

  47. [55]

    Comparison of Zigbee, Z-Wave, Wi-Fi, and Bluetooth Wireless Technologies Used in Home Automation

    Salim Jibrin Danbatta and Asaf Varol. “Comparison of Zigbee, Z-Wave, Wi-Fi, and Bluetooth Wireless Technologies Used in Home Automation”. In:2019 7th International Symposium on Digital Forensics and Security (ISDFS). 2019, pp. 1–5.DOI: 10.1109/ISDFS. 2019.8757472

    2019

  48. [56]

    Comparative analysis and practical implementation of the ESP32 microcontroller module for the internet of things

    Alexander Maier, Andrew Sharp, and Yuriy Vagapov. “Comparative analysis and practical implementation of the ESP32 microcontroller module for the internet of things”. In:2017 Internet Technologies and Applica- tions (ITA). 2017, pp. 143–148.DOI: 10.1109/ITECHA. 2017.8101926

    2017

  49. [57]

    A Multi-Protocol IoT Gateway and WiFi/BLE Sensor Nodes for Smart Home and Building Automation: Design and Implementation

    Kanitkorn Khanchuea and Rawat Siripokarpirom. “A Multi-Protocol IoT Gateway and WiFi/BLE Sensor Nodes for Smart Home and Building Automation: Design and Implementation”. In:2019 10th Interna- tional Conference of Information and Communication Technology for Embedded Systems (I...

    2019

  50. [58]

    Choice of effective messaging protocols for IoT systems: MQTT, CoAP, AMQP and HTTP

    Nitin Naik. “Choice of effective messaging protocols for IoT systems: MQTT, CoAP, AMQP and HTTP”. In:2017 IEEE International Systems Engineering Sym- posium (ISSE). 2017, pp. 1–7.DOI: 10.1109/SysEng. 2017.8088251

    2017

  51. [59]

    The Use of MQTT in M2M and IoT Systems: A Survey

    Biswajeeban Mishra and Attila Kertesz. “The Use of MQTT in M2M and IoT Systems: A Survey”. In:IEEE Access8 (2020), pp. 201071–201086.DOI: 10.1109/ ACCESS.2020.3035849

    2020

  52. [60]

    The Use and Performance of MQTT and CoAP as Internet of Things Application Protocol using NodeMCU ESP8266

    Sandy Suryo Prayogo, Yulisdin Mukhlis, and Bayu Kumoro Yakti. “The Use and Performance of MQTT and CoAP as Internet of Things Application Protocol using NodeMCU ESP8266”. In:2019 Fourth Inter- national Conference on Informatics and Computing (ICIC). 2019, pp. 1–5.DOI: 10.1109/...

    2019

  53. [61]

    On the performance of ZigBee Pro and ZigBee IP in IEEE 802.15.4 networks

    Mirko Franceschinis et al. “On the performance of ZigBee Pro and ZigBee IP in IEEE 802.15.4 networks”. In:2013 IEEE 9th International Conference on Wireless and Mobile Computing, Networking and Communica- tions (WiMob). 2013, pp. 83–88.DOI: 10.1109/WiMOB. 2013.6673344

    2013

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.