SERVICE DETAILS
Vosk STT Service
Asynchronous
Overview
The Vosk STT service is an open-source speech recognition system that converts speech to a text file.
Deployment requirements
- Memory
- 3 GiB
- Processors
- 2 vCPU0 GPU
Files & artifacts
- fdl.yml (opens in a new tab)text/yaml
OSCAR Service Definition
Show content of fdl.yml
- script.sh (opens in a new tab)text/x-shellscript
OSCAR Service Script
Show content of script.sh
- icon.png (opens in a new tab)image/png
OSCAR Service Icon
Show content of icon.png
- input_en.wav (opens in a new tab)audio/wav
Sample audio input — Replace with the audio file so that it can be processed by vosk-stt model.
- Expected text outputExpected artifacttext/plain
Expected text output — Text file produced by the vosk-stt service containing the text recognized in the audio file.
Usage & acceptance tests
Asynchronous Vosk-stt acceptance test
Upload the sample audio file, wait for the audio file to be processed, and then download the text file
Upload the sample audio file to the service's storage.
oscar-cli service put-file vosk-stt {source}Wait for the text file containing the audio recognition to appear.
Time required: PT60SDownload the latest text output file
oscar-cli service get-file vosk-stt --download-latest-into {destination}
RO-Crate metadata
- application/json
Show content of ro-crate-metadata.json