Podcastle

Podcastle is an AI audio and podcast creation platform whose developer engine, Async, exposes a low-latency Voice API for human-like text-to-speech, a browsable voice library, and instant voice cloning from a short audio sample. The API is served from https://api.async.com and authenticated with an x-api-key header plus a version header. Transcription is a Podcastle platform feature; no standalone transcription endpoint is documented in the public Voice API.

4 APIs 0 Features

AIAudioText to SpeechVoice CloningPodcasting

APIs

Podcastle Text-to-Speech API

Low-latency, human-like speech synthesis via the Async engine. Batch (/text_to_speech), streaming (/text_to_speech/streaming), word-timestamped (/text_to_speech/with_timestamps)...

Podcastle Voices API

Browse the voice library of predefined and custom voices via POST /voices, with filters for model, language, accent, gender, and style plus cursor-based pagination.

Podcastle Voice Cloning API

Instant voice cloning via POST /voices/clone. Upload a short audio sample (wav, mp3, flac, aiff) as multipart/form-data to produce a production-ready custom voice with no traini...

Podcastle Transcription

Transcription and speech-to-text are Podcastle platform features (multi-language transcription with speaker identification) available in the product, console, and credit-based p...

Collections

Async (Podcastle) Voice API

OPEN

Pricing Plans

Podcastle Plans Pricing

6 plans

PLANS

Rate Limits

Podcastle Rate Limits

3 limits

RATE LIMITS

FinOps

Podcastle Finops

FINOPS

Resources

Sources

opencollection: 1.0.0
info:
  name: Async (Podcastle) Voice API
  version: v1
request:
  auth:
    type: apikey
    apikey:
      key: x-api-key
      value: '{{apiKey}}'
      in: header
items:
- info:
    name: Text to Speech
    type: folder
  items:
  - info:
      name: Generate audio from the input text.
      type: http
    http:
      method: POST
      url: https://api.async.com/text_to_speech
      body:
        type: json
        data: "{\n  \"model_id\": \"async_flash_v1.0\",\n  \"transcript\": \"Hello from Podcastle and Async.\",\n  \"voice\"\
          : { \"mode\": \"id\", \"id\": \"<voice_id>\" },\n  \"output_format\": { \"container\": \"mp3\", \"sample_rate\"\
          : 44100, \"bit_rate\": 192000 }\n}"
    docs: Converts a full transcript into a single audio file. Only async_flash_v1.0 is documented as supported on this endpoint.
  - info:
      name: Stream audio while it is being generated.
      type: http
    http:
      method: POST
      url: https://api.async.com/text_to_speech/streaming
      body:
        type: json
        data: "{\n  \"model_id\": \"async_flash_v1.5\",\n  \"transcript\": \"Streaming low-latency speech.\",\n  \"voice\"\
          : { \"mode\": \"id\", \"id\": \"<voice_id>\" },\n  \"output_format\": { \"container\": \"mp3\", \"sample_rate\"\
          : 44100 }\n}"
    docs: Streams synthesized audio as binary chunks. Supports async_pro_v1.0, async_flash_v1.5, and async_flash_v1.0.
  - info:
      name: Generate audio with word-level timestamps.
      type: http
    http:
      method: POST
      url: https://api.async.com/text_to_speech/with_timestamps
      body:
        type: json
        data: "{\n  \"model_id\": \"async_flash_v1.0\",\n  \"transcript\": \"Words with timing.\",\n  \"voice\": { \"mode\"\
          : \"id\", \"id\": \"<voice_id>\" },\n  \"output_format\": { \"container\": \"mp3\", \"sample_rate\": 44100 }\n}"
    docs: Returns base64-encoded audio plus a word-level alignment object. Only async_flash_v1.0 is documented as supported.
- info:
    name: Voices
    type: folder
  items:
  - info:
      name: List voices from the voice library.
      type: http
    http:
      method: POST
      url: https://api.async.com/voices
      body:
        type: json
        data: "{\n  \"limit\": 10,\n  \"language\": \"en\",\n  \"gender\": \"Female\",\n  \"my_voice\": false\n}"
    docs: Returns a paginated list of predefined and custom voices with optional filters for model, language, accent, gender,
      and style.
  - info:
      name: Create an instant voice clone.
      type: http
    http:
      method: POST
      url: https://api.async.com/voices/clone
      body:
        type: multipart-form
        data: []
    docs: 'Creates a production-ready custom voice from a short audio sample (multipart/form-data). Supported formats: wav,
      mp3, flac, aiff.'
bundled: true