Skip to content

flow.injectAudio ​

Supply a local audio fixture as Chromium's microphone input. This command is available in Flowstride and Flowstride Pro; your application performs any speech-to-text processing. Follow the audio-to-text guide for a complete recording and transcript example.

Example ​

flow
flow.injectAudio "fixtures/greeting.wav";

Behavior and limits ​

Use Chromium with a project-local, uncompressed PCM WAV fixture. Firefox/WebKit audio-file injection is not supported by this command. Paths are relative to the project root, including when the flow lives in flows/.

The worker pre-scans the batch for the first literal audio path and configures the browser before launch. Flow variables and environment placeholders in this filename are not expanded by that scan. The command does not start the application's recorder or hot-swap files in a running browser. Run different fixtures in separate invocations.

sh
npx flowstride install chromium
npx flowstride run flows/audio-to-text.flow --browser chromium

Use your application's start/stop controls and microphone permission, then assert its result with transcript or scoped text checks. When using flow.waitForPipeline, the WAV fixture must also suit that helper's duration calculation.

Use this command inside a complete Given/When/Then scenario. See the browser command index for related actions.

Flowstride is MIT licensed. Flowstride Cloud offers free and paid workspace plans.