OpenClaw Skills

Heard publishes two API skills on ClawHub that let any OpenClaw developer tap into Heard’s intelligence:

`heard-transcribe`

Converts raw audio into high-fidelity, speaker-diarized text.

Input

Parameter	Type	Required	Description
`audio_url`	string	✅	URL to the audio file
`encoding`	string		Audio format (wav, flac, mp3). Default: auto-detect
`language_code`	string		BCP-47 language code. Default: `en-US`

Output

Field	Type	Description
`transcript_text`	string	Full transcript
`speaker_segments`	array	Diarized segments with speaker labels, timestamps, and text
`duration_seconds`	float	Total audio duration processed (for billing)

Processing

Asynchronous. Returns a job_id immediately. Fires a webhook to your callback URL when processing completes.

# Submit
curl -X POST https://api.heard.ai/v1/ingest/audio \
  -H "X-API-Key: sk_live_abc123" \
  -d '{"audio_url": "https://storage.example.com/call.wav"}'

# Response
{"job_id": "job_abc123", "status": "processing"}

# Poll
curl https://api.heard.ai/v1/ingest/jobs/job_abc123 \
  -H "X-API-Key: sk_live_abc123"

Pricing

Competitive per-minute pricing. Purpose-built for high volume at low cost.

`heard-analyze`

Takes a transcript and returns structured conversation intelligence. Optionally accepts customer context for richer, more strategic output.

Input

Parameter	Type	Required	Description
`transcript_text`	string	✅	The conversation transcript
`context.customer_name`	string		Customer name
`context.relationship_history`	string		Summary of previous interactions
`context.company_rules`	string		Relevant business rules or playbook excerpts
`context.custom_fields`	object		Any additional structured data (ARR, contract dates, product tier)
`metadata.participants`	array		Participant names and roles

Output

Field	Type	Description
`summary`	string	AI-generated conversation summary
`sentiment`	object	Overall and per-speaker sentiment scores
`risk_score`	float	0.0–1.0, likelihood of churn or escalation
`opportunity_score`	float	0.0–1.0, likelihood of upsell or expansion
`action_items`	array	Specific recommended next steps
`categories`	array	Detected conversation categories (billing, support, sales, etc.)

Processing

Synchronous for short transcripts (<5K tokens). Async with webhook for longer transcripts.

curl -X POST https://api.heard.ai/v1/analyze \
  -H "X-API-Key: sk_live_abc123" \
  -d '{
    "transcript_text": "Agent: Hi, how can I help today?\nCustomer: I want to cancel...",
    "context": {
      "customer_name": "Acme Corp",
      "custom_fields": {"arr": 120000, "contract_end": "2026-09-01"}
    }
  }'

Pricing

Per-analysis pricing based on transcript length. Volume discounts available.