fal-audio

You can convert text to speech via command line by running a shell script that calls the fal.ai API using curl, passing input text and a specific TTS model to generate natural-sounding speech. The script exposes model selection and text parameters as CLI arguments.

Yes, you can transcribe speech to text by providing a public audio URL to a CLI script. The script sends the URL to the fal.ai speech-to-text API via curl, which processes the audio file and returns a textual transcript suitable for chatbots and accessibility workflows.

You only need the curl command-line tool installed to execute the text-to-speech and speech-to-text scripts. The scripts rely entirely on curl to handle HTTP requests to the fal.ai API, requiring no additional programming environments or audio processing libraries.

The best way to add voice synthesis to a multimedia workflow is using a CLI script to request on-demand audio generation from fal.ai models. This approach allows you to automate text-to-speech synthesis and speech-to-text transcription by passing parameters directly through shell commands within your existing pipeline.

Yes, the speech-to-text solution works with fal.ai models and allows explicit model selection through CLI parameters. For text-to-speech, you can specify models such as fal-ai/minimax/speech-2.6-turbo to control the voice synthesis output directly from the command line.