Gradium Python Client

Python client library for the Gradium Voice AI API.

Examples

See the examples/ directory for usage examples:

• Text-to-Speech Streaming: examples/tts_streaming.py
• Speech-to-Text Streaming: examples/stt_streaming.py

You can also try it out in Google Colab.

Command Line Interface

The package includes a CLI that can be run via uvx gradium:

# Text-to-speech
uvx gradium tts "Hello world" -o output.wav
uvx gradium tts "Hello world" --voice-id abc123 -o output.ogg

# Speech-to-text
uvx gradium stt audio.wav
uvx gradium stt audio.wav --json  # outputs JSON with timestamps

# To handle non-wav file, additional dependencies are required
uvx --from 'gradium[cli]' gradium stt audio.mp3

Run uvx gradium --help for all available options.

Text-to-Speech Streaming Example

You can find some examples in the examples/ directory. Try them out with uv. For the text-to-speech streaming example, run:

uv run --with sphn examples/tts_streaming.py \
    --text "Hello, this is a test of the Gradium TTS streaming API." \
    --api-key gsk_...

Speech-to-Text Streaming Example

You can also try the speech-to-text streaming example:

uv run --with sphn examples/stt_streaming.py \
    --audio test_file.mp3 \
    --api-key gsk_...