Audio

Audio transcription and speech synthesis endpoints.

Transcriptions (Speech-to-Text)

Endpoint

POST /v1/audio/transcriptions

Request

curl https://abc123.predictor.sh/v1/audio/transcriptions \
  -H "Authorization: Bearer $TOKEN" \
  -F "[email protected]" \
  -F "model=whisper-1"

Parameters

Parameter	Type	Required	Description
file	file	Yes	Audio file (mp3, wav, m4a, webm, etc.)
model	string	Yes	Model ID (use "whisper-1")
language	string	No	Language code (e.g., "en", "es")

Response

{
  "text": "Hello, this is a transcription of the audio file."
}

Supported Audio Formats

• WAV (.wav)

• MP3 (.mp3)

• M4A/AAC (.m4a, .aac)

• OGG Vorbis (.ogg, .oga)

• FLAC (.flac)

• WebM (.webm)

Python Example

from openai import OpenAI

client = OpenAI(
    base_url="https://abc123.predictor.sh/v1",
    api_key="pred_your_token"
)

with open("meeting.mp3", "rb") as audio:
    transcript = client.audio.transcriptions.create(
        model="whisper-1",
        file=audio,
        language="en"
    )
print(transcript.text)

---

Speech (Text-to-Speech)

OpenAI-Compatible Endpoint

POST /v1/audio/speech

Request

{
  "input": "Hello, this is a test of text to speech.",
  "voice": "alloy",
  "response_format": "mp3",
  "speed": 1.0
}

Parameters

Parameter	Type	Required	Description
input	string	Yes	Text to synthesize
voice	string	Yes	Voice ID
response_format	string	No	mp3 (default) or wav
speed	float	No	Speed 0.5 to 2.0 (default 1.0)
stream_format	string	No	Set to sse for streaming

Available Voices

Voice	Description
alloy	Neutral female
echo	Neutral male
fable	British male
onyx	Deep male
nova	Energetic female
shimmer	Soft female

Response

Binary audio data (MP3 or WAV).

Example

curl https://abc123.predictor.sh/v1/audio/speech \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "input": "Hello world!",
    "voice": "alloy"
  }' \
  --output speech.mp3

Python Example

from openai import OpenAI

client = OpenAI(
    base_url="https://abc123.predictor.sh/v1",
    api_key="pred_your_token"
)

response = client.audio.speech.create(
    model="tts-1",
    voice="alloy",
    input="Hello, this is a test.",
    speed=1.0
)

response.stream_to_file("output.mp3")

---

ElevenLabs-Compatible Endpoints

Generate Speech

POST /v1/text-to-speech/{voice_id}

curl https://abc123.predictor.sh/v1/text-to-speech/af_bella \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"text": "Hello world"}' \
  --output speech.mp3

Streaming Speech

POST /v1/text-to-speech/{voice_id}/stream

curl https://abc123.predictor.sh/v1/text-to-speech/af_bella/stream \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"text": "Hello world"}'

Response is Server-Sent Events with base64-encoded audio chunks.

List Voices

GET /v1/voices

curl https://abc123.predictor.sh/v1/voices \
  -H "Authorization: Bearer $TOKEN"

Response:

{
  "voices": [
    {"voice_id": "af_bella", "name": "Bella"},
    {"voice_id": "am_adam", "name": "Adam"},
    ...
  ]
}

Voice IDs (Kokoro)

Voice ID	Description
af_bella	American female, warm
af_sarah	American female, professional
af_nicole	American female, energetic
af_sky	American female, soft
am_adam	American male, neutral
am_michael	American male, deep
bf_emma	British female, warm
bf_isabella	British female, elegant
bm_george	British male, classic
bm_lewis	British male, friendly