TinyML
STFT-TCNN speech enhancement
Frequency-domain temporal convolutional network that masks noise in speech spectrograms. Adapted by STMicroelectronics for int8 use on STM32N6.
Levelmodel
ClassTinyML
Parametersnot in the notes
Contextnot in the notes
Releasednot in the notes
Opennessunknown
Runs onmicrocontroller
Statusactive
R&D Topics
Identity
- Developer
- STMicroelectronics
- Family
- not in the notes
- Released
- – sources
- Weights published
- –
- Record checked
- 12 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.
Architecture and Size
- Architecture
- CNN TCN core of Pandey and Wang TCNN with STFT pre-processing and inverse STFT post-processing so it quantises to int8
- Total parameters
- – sources
- Active parameters
- –
- Checkpoint parameters
- –
- Context length
- not in the notes
- Input
- audio
- Output
- audio
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).
Openness and Access
- Openness
- unknown
- Licence
- not in the notes sources
- Access routes
- open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.
Deployment
- Runs on
- microcontroller
- Footprint
- README benchmarks on STM32N6 (STM32N6570-DK).
- Runtimes
- not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.
Indonesia
- Indonesian support
- not stated
- Regional languages
- not in the notes
- Jakarta hosting
- unknown
- Languages
- not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.
Benchmark Results
The notes record no benchmark result for this model.
Lineage
The notes record no lineage for this model.
Industry Use Cases
No industry use case links this model.
Links
ModelSTFT-TCNN speech enhancementSTMicroelectronics, TinyML
STFT-TCNN speech enhancementSTMicroelectronics, TinyMLUse Casesnone recorded
Who Runs Itnone recorded
Organisations 1
Sources
- github.com/STMicroelectronics/stm32ai-modelzoo/tree/main... https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/speech_enhancement
- github.com/STMicroelectronics/stm32ai-modelzoo/tree/main... https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/speech_enhancement/stft_tcnn
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "STFT-TCNN Speech Enhancement", https://aurion.semai.id/models/stmicroelectronics-stft-tcnn/, records of October 2026.