Pipeline Templates

Pre-configured voice pipelines for common use cases. Templates combine STT, LLM, and TTS with optimized settings.

Available Templates

Template	Use Case	Latency	Quality
voice-agent-fast	Real-time voice assistant	Low	Good
voice-agent-premium	High-quality voice assistant	Medium	Excellent
ambient-scribe	Medical transcription	Low	Excellent
legal-dictation	Legal document dictation	Medium	Excellent

voice-agent-fast

Optimized for low-latency voice interactions.

template: voice-agent-fast
stt:
  model: voxtral-mini-latest
  streaming: true
llm:
  model: gpt-5-mini
  max_tokens: 150
  temperature: 0.7
tts:
  model: eleven_turbo_v2_5
  voice: friendly
  streaming: true

Usage:

curl -X POST https://api.gateflow.ai/v1/audio/pipelines \
  -H "Authorization: Bearer gw_prod_..." \
  -F "audio=@question.mp3" \
  -F "template=voice-agent-fast" \
  -F "context=You are a helpful assistant."

Best for:

• Voice assistants
• Customer service bots
• Interactive voice response (IVR)

voice-agent-premium

Maximum quality for premium voice experiences.

template: voice-agent-premium
stt:
  model: whisper-1
  streaming: false
llm:
  model: gpt-5.2
  max_tokens: 500
  temperature: 0.7
tts:
  model: eleven_multilingual_v2
  voice: professional
  streaming: true

Usage:

curl -X POST https://api.gateflow.ai/v1/audio/pipelines \
  -H "Authorization: Bearer gw_prod_..." \
  -F "audio=@question.mp3" \
  -F "template=voice-agent-premium"

Best for:

• Executive assistants
• Premium customer experiences
• Accessibility applications

ambient-scribe

Medical transcription with HIPAA-compliant PII handling.

template: ambient-scribe
stt:
  model: whisper-1
  language: en
  word_timestamps: true
llm:
  model: gpt-5.2
  system_prompt: |
    Extract and structure medical information from the transcript.
    Identify: chief complaint, symptoms, medications, allergies.
    Format as structured clinical note.
pii:
  detect: true
  redact: false
  categories: [PHI, medications, diagnoses]
compliance:
  hipaa: true
  audit_log: true

Usage:

curl -X POST https://api.gateflow.ai/v1/audio/pipelines \
  -H "Authorization: Bearer gw_prod_..." \
  -F "audio=@patient_visit.mp3" \
  -F "template=ambient-scribe"

Response:

{
  "transcription": "Patient reports headache for 3 days...",
  "structured_note": {
    "chief_complaint": "Headache",
    "duration": "3 days",
    "symptoms": ["headache", "fatigue"],
    "medications_mentioned": ["ibuprofen"]
  },
  "pii_detected": [
    {"type": "patient_name", "redacted": false}
  ],
  "audit_id": "audit_abc123"
}

Best for:

• Medical dictation
• Clinical note generation
• Healthcare documentation

legal-dictation

Legal document transcription with citation detection.

template: legal-dictation
stt:
  model: whisper-1
  language: en
llm:
  model: gpt-5.2
  system_prompt: |
    Transcribe and format as legal document.
    Identify case citations, statutes, and legal terms.
    Maintain proper legal formatting.
formatting:
  style: legal
  citations: detect
  paragraphs: numbered

Usage:

curl -X POST https://api.gateflow.ai/v1/audio/pipelines \
  -H "Authorization: Bearer gw_prod_..." \
  -F "audio=@deposition.mp3" \
  -F "template=legal-dictation"

Best for:

• Deposition transcription
• Legal memo dictation
• Court filing preparation

Custom Templates

Create your own templates:

curl -X POST https://api.gateflow.ai/v1/management/pipeline-templates \
  -H "Authorization: Bearer gw_prod_admin_key" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "my-custom-template",
    "stt": {
      "model": "whisper-1"
    },
    "llm": {
      "model": "gpt-5.2",
      "system_prompt": "You are a helpful assistant.",
      "max_tokens": 200
    },
    "tts": {
      "model": "eleven_turbo_v2_5",
      "voice": "friendly"
    }
  }'

Template Overrides

Override template settings per request:

response = requests.post(
    "https://api.gateflow.ai/v1/audio/pipelines",
    headers={"Authorization": "Bearer gw_prod_..."},
    json={
        "template": "voice-agent-fast",
        "audio": base64_audio,
        "overrides": {
            "llm": {
                "model": "claude-sonnet-4-5-20250929",
                "max_tokens": 300
            },
            "tts": {
                "voice": "professional"
            }
        }
    }
)

Next Steps

• Streaming Speech - Real-time audio streaming
• Audio Providers - Provider configuration