strands-deepgram

Deepgram speech processing tool for Strands Agents SDK. Enables powerful speech-to-text, text-to-speech, and audio intelligence capabilities for AI agents.

Features

• Speech-to-Text: Transcribe audio with 30+ language support
• Text-to-Speech: Generate natural-sounding speech in multiple voices
• Audio Intelligence: Sentiment analysis, topic detection, intent recognition
• Speaker Diarization: Identify and separate different speakers
• Multi-format Support: WAV, MP3, M4A, FLAC, and more
• Type Safe: Full type hints and validation
• Easy Integration: Drop-in tool for Strands agents

Requirements

• Python 3.9+
• Strands Agents SDK 1.11.0+
• Deepgram SDK 3.0+

Installation

pip install strands-deepgram

Quick Start

from strands import Agent
from strands_deepgram import deepgram

# Create an agent with Deepgram tool
agent = Agent(tools=[deepgram])

# Transcribe audio with speaker identification
agent("transcribe audio from recording.mp3 in Turkish with speaker diarization")

# Text-to-speech
agent("convert this text to speech and save as output.mp3: Hello world")

# Audio intelligence
agent("analyze sentiment and topics in recording.wav")

Configuration

Set your Deepgram API key as an environment variable:

DEEPGRAM_API_KEY=your_deepgram_api_key  # Required
DEEPGRAM_DEFAULT_MODEL=nova-3            # Optional
DEEPGRAM_DEFAULT_LANGUAGE=en             # Optional

Get your API key at: console.deepgram.com

Supported Actions

Speech-to-Text (transcribe)

agent("transcribe this audio file: path/to/audio.mp3")

Features:

• Multi-language transcription (30+ languages)
• Speaker diarization (identify different speakers)
• Smart formatting and punctuation
• Word-level timestamps
• Sentiment analysis (optional)
• Topic and intent detection (optional)

Text-to-Speech (text_to_speech)

agent("convert this text to speech: Hello, how are you today?")

Features:

• Natural-sounding voices (Aura series)
• Multiple audio formats (MP3, WAV, FLAC)
• Customizable speech parameters
• Voice selection

Audio Intelligence (analyze)

agent("analyze sentiment and topics in audio: call.mp3")

Features:

• Sentiment analysis
• Topic detection
• Intent recognition
• Language detection

Testing

# Install development dependencies
pip install -e ".[dev]"

# Run tests
pytest tests/ -v

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Links

• PyPI: pypi.org/project/strands-deepgram
• GitHub: github.com/eraykeskinmac/strands-deepgram
• Strands Agents SDK: github.com/strands-agents/strands
• Deepgram API: developers.deepgram.com

Built for the Strands community 🚀