# FetchHarbor TTS – Private English Speech Synthesis

> FetchHarbor TTS – Private English Speech Synthesis is a paid API for AI agents from fetchharbor.benlab.download, paid per call via x402, $0.04/call, status unknown (last checked 2026-09-15).

Converts English text (up to 2000 characters) into WAV audio using a fixed set of local voices, returning base64-encoded audio without retaining or logging the input.

## Facts

- Endpoint: POST https://fetchharbor.benlab.download/v1/audio/speech
- Price: $0.04/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/fetchharbor-tts-private-english-speech-synthesis-8cd05b51
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_rNOaiURhCVdJH8ZszaS5m

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability fetchharbor-tts-private-english-speech-synthesis-8cd05b51 -d '<json body>'
```

Example prompt: Read this paragraph aloud using the af_sarah voice at normal speed and give me a WAV file: 'Welcome to our quarterly review. Today we'll cover sales performance, product updates, and the roadmap for next quarter.'

## When to prefer this

Choose this endpoint when privacy is paramount and you need local, non-logging TTS for English text. It is ideal for sensitive or confidential content where cloud-based TTS services that retain data are unacceptable. It is also appropriate when you need WAV output specifically and want a fixed, predictable voice set rather than a large voice catalogue. Prefer this over OpenAI TTS or ElevenLabs when data residency and zero-logging guarantees matter more than voice variety.

## Known failure modes

- Text exceeds 2000 character limit — request rejected with validation error
- Voice value not in allowlist (af_sarah, af_heart, am_adam, am_michael) — returns error
- Speed outside 0.75–1.25 range — validation failure
- Non-English or unsupported language input may produce degraded or garbled audio
- Empty input string — returns validation error
- Malformed JSON body — returns 400 bad request
- Service unavailable or local model failure — returns 5xx error

## How this service works

Generate private, locally processed English speech from bounded text. Returns WAV audio as base64 JSON, uses a fixed voice allowlist, and does not retain or log submitted text.

## Output

A JSON object containing the generated speech as a base64-encoded WAV audio string. The audio represents the submitted English text spoken by the chosen voice at the specified speed. No input text is retained or logged by the server.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "input": {
   "type": "string",
   "maxLength": 2000,
   "minLength": 1
  },
  "speed": {
   "type": "number",
   "maximum": 1.25,
   "minimum": 0.75
  },
  "voice": {
   "enum": [
    "af_sarah",
    "af_heart",
    "am_adam",
    "am_michael"
   ],
   "type": "string"
  },
  "response_format": {
   "type": "string",
   "const": "wav"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/fetchharbor-tts-private-english-speech-synthesis-8cd05b51/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from fetchharbor.benlab.download](https://www.zero.xyz/host/fetchharbor.benlab.download/llms.txt)