Skip to main content
Agents automatically handle audio files when included in prompts. First validate your audio file with Timbal’s File type, then pass it alongside text in a list: 
from timbal import Agent
from timbal.types.file import File

agent = Agent(
    name="AudioAgent",
    model="openai/gpt-4o-audio-preview", # Audio-capable model
    system_prompt="Transcribe and analyze audio content."
)

# Validate audio file and analyze
audio_file = File.validate("path/to/recording.mp3")
result = await agent(
    prompt=["Transcribe this audio and summarize the key points", audio_file]
).collect()

print(result.output.content[0].text)
Use vision-capable models for audio processing. Check Model Capabilities to see which models support audio input.

​
Key Features

  • Automatic Processing: Audio files are automatically transcribed to text
  • Speech Analysis: Extract content, sentiment, and insights from audio recordings
  • File Support: Works with local files, URLs, and base64 data