Built by Metorial, the integration platform for agentic AI.
Provider Summary
transcribe pre-recorded audio
transcribe live streaming audio
convert text to speech
detect speakers in audio
analyze text sentiment
detect topics and intents
summarize transcripts
build conversational voice agents
manage projects and keys
discover available models
Transcribe pre-recorded and live streaming audio to text in 45+ languages with speaker diarization, smart formatting, and keyword boosting. Convert text to natural-sounding speech with 40+ voice options. Analyze transcripts for sentiment, topics, summaries, and intents. Build conversational voice agents with integrated STT, LLM reasoning, and TTS in a single session. Manage projects, API keys, members, billing, and usage. Discover available models and their metadata. Supports asynchronous processing via callbacks for both transcription and speech synthesis.
Analyze text for intelligence insights including sentiment analysis, topic detection, intent detection, and summarization. Enable one or more analysis features to extract value from text content such as transcripts, articles, or conversations.
Get usage data for a Deepgram project. Filter by date range, API key, tag, method (sync/async/streaming), or model. Useful for monitoring API consumption and billing.
Query available Deepgram models and their metadata. Useful for discovering which models are available for transcription or text-to-speech and what languages they support.
List all API keys for a Deepgram project. Returns key metadata including comments, scopes, tags, and expiration dates. Does not return the actual key values.
List all members of a Deepgram project. Returns member details including name, email, and permission scopes.
List all Deepgram projects accessible with the current API key. Returns project IDs, names, and company information.
Convert text into natural-sounding speech audio. Returns base64-encoded audio data. Supports 40+ English voices with localized accents, configurable encoding formats, sample rates, and bit rates.
Transcribe pre-recorded audio to text. Supports audio from a URL or raw audio data (base64-encoded). Provides options for model selection, language detection, speaker diarization, smart formatting, keyword boosting, and text intelligence features (summarization, topic detection, sentiment analysis). Returns the full transcript with word-level timestamps and confidence scores.
This integration is licensed under the AGPL-3.0 License.
Built with ❤️ by Metorial