Edge Impulse

Edge Impulse DSP API

The DSP API from Edge Impulse — 18 operation(s) for dsp.

GitHub OpenAPI
Business capability
Artificial Intelligence Management BC-610.60

Operations 19

Each operation below carries the questions people ask an LLM about it and the instructions they give an agent to run it. Generated by API Evangelist overlay

GET /api/{projectId}/dsp/{dspId} Get a DSP block's configuration · Get config #
Ask an LLM
“How do I read the current parameters of a DSP processing block in my impulse?”
“What settings is my spectral or MFCC block using right now?”
Tell an agent
Get the configuration of DSP block {dspId} in project {projectId}.
Show the processing parameters for DSP block {dspId}.
POST /api/{projectId}/dsp/{dspId} Update a DSP block's configuration · Set config #
Ask an LLM
“Can I change only some DSP block parameters without overwriting the rest?”
“How do I save new processing parameters to a DSP block?”
Tell an agent
Set the config of DSP block {dspId} in project {projectId} to {config}.
Save parameters {config} on DSP block {dspId}.
POST /api/{projectId}/dsp/{dspId}/clear Clear a DSP block's generated features · Clear DSP block #
Ask an LLM
“How do I wipe the features a DSP block has already generated?”
“Can I reset a processing block so its features have to be regenerated?”
Tell an agent destructive · confirm first
Clear the generated features of DSP block {dspId} in project {projectId}.
Reset DSP block {dspId} by discarding its features.
GET /api/{projectId}/dsp/{dspId}/metadata Get metadata from a generated DSP block · Get metadata #
Ask an LLM
“What metadata does a DSP block record after features are generated?”
“Can I leave the list of included samples out of the DSP metadata?”
Tell an agent
Get the generation metadata for DSP block {dspId} in project {projectId}.
Fetch DSP block {dspId} metadata without the included samples list.
GET /api/{projectId}/dsp/{dspId}/raw-data/{sampleId} Get a raw sample filtered to the DSP block's axes · Get raw sample #
Ask an LLM
“How do I view a full raw sample showing only the axes my DSP block uses?”
“Can I cap how many payload values come back for a raw sample?”
Tell an agent
Get raw sample {sampleId} in project {projectId} filtered to the axes of DSP block {dspId}.
Show the whole of sample {sampleId} for DSP block {dspId}, limited to {limitPayloadValues} values.
GET /api/{projectId}/dsp/{dspId}/raw-data/{sampleId}/slice Get a slice of raw sample data for a DSP block · Get raw sample (slice) #
Ask an LLM
“Can I fetch just a time window of raw data from a sample, before processing?”
“How do I get the unprocessed raw values between two offsets of a sample?”
Tell an agent
Get the raw slice of sample {sampleId} in project {projectId} from {sliceStart} to {sliceEnd} for DSP block {dspId}.
Show unprocessed raw data of sample {sampleId} starting at {sliceStart}.
POST /api/{projectId}/dsp/{dspId}/raw-data/{sampleId}/slice/run Process a sample slice with custom parameters · Get processed sample (slice) #
Ask an LLM
“How do I preview what a DSP block outputs for a window using parameters I haven't saved?”
“Can I run a slice through processing and store the result?”
Tell an agent
Run the slice of sample {sampleId} in project {projectId} starting at {sliceStart} through DSP block {dspId} with parameters {params}.
Process sample {sampleId} from {sliceStart} to {sliceEnd} with params {params} and store it as {store}.
GET /api/{projectId}/dsp/{dspId}/raw-data/{sampleId}/slice/run/readonly Process a sample slice with saved parameters · Get processed sample (slice) #
Ask an LLM
“Is there a read-only way to see processed output for a slice using the block's saved settings?”
“Can I view processed features for a window without changing anything?”
Tell an agent
Show the processed output of sample {sampleId} in project {projectId} from {sliceStart} using DSP block {dspId}'s saved config, read-only.
Get read-only processed features for sample {sampleId} between {sliceStart} and {sliceEnd}.
GET /api/{projectId}/dsp/{dspId}/features/labels List the feature names a DSP block generates · Feature labels #
Ask an LLM
“What are the names of the features my processing block produces?”
“Which feature labels come out of a given DSP block?”
Tell an agent
List the feature names generated by DSP block {dspId} in project {projectId}.
Show the feature labels for DSP block {dspId}.
GET /api/{projectId}/dsp/{dspId}/features/get-graph/{category} Sample trained features for the feature explorer · Sample of trained features #
Ask an LLM
“How do I plot my training set in the feature explorer on three chosen axes?”
“Can I get a sample of trained features with labels for visualization?”
Tell an agent
Sample trained features from DSP block {dspId} for the {category} set on axes {featureAx1}, {featureAx2} and {featureAx3}.
Get a feature explorer graph for project {projectId} block {dspId} using axes {featureAx1}, {featureAx2}, {featureAx3}.
GET /api/{projectId}/dsp/{dspId}/features/get-graph/classification/{sampleId} Generate features for every window of a sample · Features for sample #
Ask an LLM
“How do I see the features a DSP block produces as the sliding window moves across a whole sample?”
“Can I get per-window features for one sample to plot it in the feature explorer?”
Tell an agent
Run DSP block {dspId} across all windows of sample {sampleId} and return the features.
Get sliding-window features of sample {sampleId} in project {projectId} for the feature graph.
GET /api/{projectId}/dsp/{dspId}/features/importance Get feature importance for a DSP block · Feature importance #
Ask an LLM
“Which features from my DSP block matter most to the model?”
“Why is feature importance unavailable when dimensionality reduction is on?”
Tell an agent
Get feature importance for DSP block {dspId} in project {projectId}.
Rank the features of DSP block {dspId} by importance.
GET /api/{projectId}/dsp-data/{dspId}/x/{category} Download DSP output as a NumPy array · Download DSP data #
Ask an LLM
“Can I download the processed features for my whole training set as a NumPy file?”
“How do I export windowed DSP output, or the raw windows, for offline training?”
Tell an agent
Download the DSP output (X) of block {dspId} for the {category} set.
Export raw windowed data from DSP block {dspId} in project {projectId} for {category} with raw set to {raw}.
GET /api/{projectId}/dsp-data/{dspId}/y/{category} Download labels for windowed DSP output · Download DSP labels #
Ask an LLM
“Where do I get the labels (y) that match the downloaded DSP feature windows?”
“Can I export the per-window labels of the training set for a DSP block?”
Tell an agent
Download the labels (y) for DSP block {dspId} in the {category} set.
Export window labels from project {projectId} block {dspId} for {category}.
GET /api/{projectId}/dsp/{dspId}/get-autotuner-results Get DSP autotuner parameter suggestions · Get results from DSP autotuner #
Ask an LLM
“What parameters did the DSP autotuner recommend for my block?”
“Can I see the autotuner's suggested settings before applying them?”
Tell an agent
Get the autotuner results for DSP block {dspId} in project {projectId}.
Show the parameter sets the autotuner found for block {dspId}.
GET /api/{projectId}/dsp-data/{dspId}/artifact/{key} Download a DSP debugging artifact · Download a DSP artifact #
Ask an LLM
“Can I pull a debugging artifact a DSP block produced by its key?”
“How do I fetch internal DSP artifacts to debug processing?”
Tell an agent
Download DSP artifact {key} from block {dspId}.
Fetch debug artifact {key} for DSP block {dspId} in project {projectId}.
POST /api/{projectId}/dsp/{dspId}/profile Profile a custom DSP block's performance · Profile custom DSP block #
Ask an LLM
“How do I measure latency and memory for my own custom DSP block?”
“Does profiling a custom block need a TFLite implementation, and how do I follow progress?”
Tell an agent
Start profiling custom DSP block {dspId} with parameters {params}.
Kick off a performance profile for custom block {dspId} in project {projectId} using {params}.
GET /api/{projectId}/dsp/{dspId}/performance Estimate DSP block latency on every device · Get DSP block performance for all latency devices #
Ask an LLM
“How fast will my DSP block run on each supported target device?”
“What RAM does a processing block need across all latency devices?”
Tell an agent
Get estimated latency and RAM for DSP block {dspId} on all devices.
Show per-device performance of DSP block {dspId} in project {projectId}.
POST /api/{projectId}/dsp/{dspId}/run Process a raw features array with a DSP block · Get processed sample (from features array) #
Ask an LLM
“Can I push my own raw sensor array through a DSP block without uploading a sample?”
“How do I get processed features and graphs from an ad-hoc features array?”
Tell an agent
Run features {features} through DSP block {dspId} in project {projectId} with parameters {params}.
Process array {features} with block {dspId}, drawGraphs {drawGraphs} and requestPerformance {requestPerformance}.

Documentation

📖
Documentation
https://docs.edgeimpulse.com/reference/edge-impulse-api

Specifications

⚙
OpenAPI
https://raw.githubusercontent.com/api-evangelist/edge-impulse/refs/heads/main/openapi/edge-impulse-dsp-api-openapi.yml

Other Resources

🔗
Overlay
https://raw.githubusercontent.com/api-evangelist/edge-impulse/refs/heads/main/overlays/edge-impulse-dsp-overlay.yaml

Work with this as data

Every API here is available over the APIs.io API and to AI agents over MCP.

MCP server

One button, every client — Claude, Cursor, VS Code and the rest.

https://apis.io/mcp

Tools for apis

7 MCP tools reach this
  • find_apisBrowse and filter every API in the catalog.
  • get_api_artifactsOne API's artifacts, grouped by type.
  • get_openapiThe primary OpenAPI for this API.
  • find_similar_apisAPIs that look like this one.
  • apis_io_searchSTART HERE — APIs, providers and tags for one query, each with its total.
  • resolveTurn a domain, URL or GitHub org into the provider it belongs to.
  • find_cohortsEvery scored population of providers in the catalog.
All 92 tools →

Call it yourself

curl for this page
This API 
curl "https://apis.io/api/v1/apis/edge-impulse:edge-impulse-dsp-api"
All apis 
curl "https://apis.io/api/v1/apis?limit=25"

Discovery needs no key. Ratings and market analysis are Pro.

Get an API key

Free tier, no form to fill in. Signing in shares your email address with us — we store it to create your key and to recognise you if you sign in with another provider. See our Privacy Policy and Terms.

GitHub Google LinkedIn

A second provider on the same verified email joins the account you already have.

OpenAPI Specification

edge-impulse-dsp-api-openapi.yml Raw ↑ 
openapi: 3.2.0
info:
  title: Edge Impulse DSP API
  version: 1.0.0
  description: 'Operations tagged DSP across 2 of this provider''s published API definitions: edge-impulse-openapi.yml, edge-impulse-studio-openapi-2026-09-06.yml. Each path carries the servers of the definition it was published in.'
servers:
- url: https://studio.edgeimpulse.com/v1
tags:


# --- truncated at 32 KB (81 KB total) ---
# Full source: https://raw.githubusercontent.com/api-evangelist/edge-impulse/refs/heads/main/openapi/edge-impulse-dsp-api-openapi.yml