Unscripted API demo

Use this page to test the ELSA unscripted API:

a) Begin by recording an audio file locally on your computer. For optimal analysis results, try for a duration of more than 20 seconds. If you're unsure what to say, you can download this sample audio.

b) Upload your recorded audio using the upload button below, along with your API key.

c) Submit your data by clicking on the submit button. The results will be displayed in the text box below, in JSON format.

Alternatively, you can execute the same command in a command line using cURL by copying and pasting the automatically generated command located above the submit button.

Important: Ensure that your audio is encoded in the correct format to be accepted by the service. Currently, we accept FLAC, OGG-Opus, and WAV formats. For WAV files, the audio must be in 16-bit, 16KHz, mono channel format; otherwise, the server will reject it. If your audio is in another format, you can use the ffmpeg tool or the following sox command for conversion:

sox input_audio_file -r 16000 -c 1 -b 16 audio.wav

NOTE: This version of the API may respond slowly when processing very long audio files. If your use-case involves long audio recordings and streaming is feasible, we recommend integrating with our API in streaming mode. Feel free to reach out to us for assistance.