# BaalSTT Speech-to-Text API

> BaalSTT Speech-to-Text API is a paid API for AI agents from baalstt.tailb3821a.ts.net, paid per call via x402, $0.05/call, status down (last checked 2026-10-12).

Transcribes audio from a public URL into text using Deepgram nova-2 or Whisper large-v3 models, with pay-per-call USDC pricing via x402.

## Facts

- Endpoint: POST https://baalstt.tailb3821a.ts.net/transcribe?utm_source=zero.xyz
- Price: $0.05/call
- Payment: x402
- Status: down
- Last checked: 2026-10-12
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/baalstt-speech-to-text-api-3f28154d
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_vtixc5uOyvkRezia5p87k

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 baalstt-speech-to-text-api-3f28154d -d '<json body>'
```

Example prompt: Can you transcribe the audio file at https://example.com/interview.mp3 using the whisper-large-v3 model and give me the full text?

## When to prefer this

Choose BaalSTT when you need quick, low-cost audio-to-text transcription from a public URL without account setup or KYC — especially in agent workflows where per-call micropayments via x402/USDC are preferred over monthly subscriptions. It offers a choice between Deepgram nova-2 (fast, accurate) and Whisper large-v3 (high accuracy, multilingual), making it flexible for both real-time and batch-style use cases.

## Known failure modes

- Audio URL is not publicly accessible or returns a non-200 status — transcription will fail
- Unsupported audio format at the provided URL
- Free tier (3 calls per 24h per IP) exhausted without payment — returns 402 Payment Required
- Network timeout if audio file is very large or slow to download
- Invalid model enum value provided — request rejected
- USDC payment failure or insufficient balance when free tier is exceeded

## How this service works

BaalSTT — Speech-to-Text API (Deepgram nova-2 / whisper-large-v3). Audio in -> text out. AI agents pay per call via x402/USDC (gasless, no KYC); 3 free transcriptions / 24 h / IP without payment.

## Output

Returns a JSON object containing the full transcribed text string, the model name used for transcription, and the duration of the audio in seconds (may be null for some models).

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "url"
 ],
 "properties": {
  "url": {
   "type": "string",
   "description": "Public http(s) URL of the audio to transcribe"
  },
  "model": {
   "enum": [
    "nova-2",
    "whisper-large-v3",
    "whisper-large-v3-turbo"
   ],
   "type": "string"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "text": {
   "type": "string",
   "description": "Transcribed text"
  },
  "model": {
   "type": "string"
  },
  "duration_s": {
   "type": [
    "number",
    "null"
   ],
   "description": "Audio duration in seconds"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/baalstt-speech-to-text-api-3f28154d/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from baalstt.tailb3821a.ts.net](https://www.zero.xyz/host/baalstt.tailb3821a.ts.net/llms.txt)