Generate Audio with Gemini

Generate audio from text using Gemini Live API.

Python

from google import genai
import asyncio
import contextlib
import wave

client = genai.Client(http_options= {
      'api_version': 'v1alpha'
})
model = "gemini-2.0-flash-exp"
audio_negative_prompt = "dont say OK , I will do this or that, just only read this story using voice expressions without introductions or ending ,more segments are comming ,dont say OK , I will do this or that:\n"

@contextlib.contextmanager
def wave_file(filename, channels=1, rate=24000, sample_width=2):
    with wave.open(filename, "wb") as wf:
        wf.setnchannels(channels)
        wf.setsampwidth(sample_width)
        wf.setframerate(rate)
        yield wf

def generate_audio_live(text, output_filename):
    collected_audio = bytearray()

    async def _generate():
        config = {
            "generation_config": {"response_modalities": ["AUDIO"]}
        }
        async with client.aio.live.connect(model=model, config=config) as session:
            await session.send(input=audio_negative_prompt + text, end_of_turn=True)
            async for response in session.receive():
                if response.data:
                    collected_audio.extend(response.data)
        return bytes(collected_audio)

    audio_bytes = asyncio.run(_generate())
    with wave_file(output_filename) as wf:
        wf.writeframes(audio_bytes)
    return output_filename