​

Audio Models Overview

​

Available Models

​

Text-to-Audio Models

​

OpenAI TTS

• TTS-1: Fast, efficient text-to-speech
• TTS-1-HD: High-quality voice synthesis

​

ElevenLabs

• Eleven Turbo: Ultra-fast voice generation
• Eleven Multilingual: Support for 29 languages
• Eleven Monolingual: Premium English voices

​

Google Cloud

• WaveNet: Neural voice synthesis
• Standard: Traditional concatenative TTS

​

Audio-to-Text Models

​

OpenAI Whisper

• Whisper Large: Highest accuracy for transcription
• Whisper Base: Balanced speed and accuracy
• Whisper Tiny: Fast processing for real-time use

​

Google Speech

• Chirp: Latest generation speech recognition
• Standard: Reliable speech-to-text conversion

​

AssemblyAI

• Universal-1: Advanced speech understanding
• Best: Highest accuracy model

​

Model Capabilities

​

Text-to-Audio API

POST /v1/audio/speech

​

Basic Example

curl -X POST "https://api.anyapi.ai/v1/audio/speech" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "tts-1",
    "input": "Hello, this is a test of text-to-speech conversion.",
    "voice": "alloy",
    "response_format": "mp3"
  }' \
  --output speech.mp3

​

Available Voices

​

OpenAI TTS Voices

• alloy: Neutral, balanced tone
• echo: Warm, engaging voice
• fable: Expressive, storytelling voice
• onyx: Deep, authoritative tone
• nova: Bright, energetic voice
• shimmer: Soft, gentle tone

​

ElevenLabs Voices

• Rachel: Professional female voice
• Drew: Conversational male voice
• Clyde: Middle-aged male voice
• Paul: Mature, authoritative male
• Domi: Confident female voice
• Dave: British male accent

​

Audio-to-Text API

POST /v1/audio/transcriptions

​

Basic Example

curl -X POST "https://api.anyapi.ai/v1/audio/transcriptions" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F file="@audio.mp3" \
  -F model="whisper-1" \
  -F language="en"

​

Response Format

{
  "text": "Hello, this is a transcription of the audio file.",
  "language": "en",
  "duration": 3.2,
  "segments": [
    {
      "id": 0,
      "start": 0.0,
      "end": 3.2,
      "text": "Hello, this is a transcription of the audio file.",
      "confidence": 0.98
    }
  ]
}

​

Translation API

POST /v1/audio/translations

import requests

with open("spanish_audio.mp3", "rb") as audio_file:
    response = requests.post(
        "https://api.anyapi.ai/v1/audio/translations",
        headers={"Authorization": "Bearer YOUR_API_KEY"},
        files={"file": audio_file},
        data={
            "model": "whisper-1",
            "source_language": "es",
            "target_language": "en"
        }
    )

print(response.json())

​

Advanced Features

​

SSML Support

{
  "model": "tts-1-hd",
  "input": "<speak><prosody rate='slow'>Hello there!</prosody> <break time='1s'/> How are you today?</speak>",
  "voice": "alloy",
  "response_format": "wav"
}

​

Voice Cloning

# Upload voice sample
files = {'voice_sample': open('voice_sample.wav', 'rb')}

response = requests.post(
    "https://api.anyapi.ai/v1/voices/clone",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    files=files,
    data={
        "name": "My Custom Voice",
        "description": "Personal voice clone"
    }
)

voice_id = response.json()['voice_id']

# Use custom voice
requests.post(
    "https://api.anyapi.ai/v1/audio/speech",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "model": "elevenlabs-turbo",
        "input": "This is my cloned voice speaking!",
        "voice": voice_id
    }
)

​

Real-time Streaming

const ws = new WebSocket('wss://api.anyapi.ai/v1/audio/stream');

ws.onopen = () => {
  ws.send(JSON.stringify({
    model: 'whisper-base',
    language: 'en',
    format: 'pcm16'
  }));
};

// Send audio chunks
navigator.mediaDevices.getUserMedia({audio: true})
  .then(stream => {
    const mediaRecorder = new MediaRecorder(stream);
    mediaRecorder.ondataavailable = (event) => {
      ws.send(event.data);
    };
    mediaRecorder.start(100); // Send chunks every 100ms
  });

ws.onmessage = (event) => {
  const result = JSON.parse(event.data);
  console.log('Transcription:', result.text);
};

​

Model Comparison

​

Text-to-Speech Models

​

Speech-to-Text Models

​

Supported Languages

​

Text-to-Speech

• English: All models
• Spanish: ElevenLabs, Google
• French: ElevenLabs, Google
• German: ElevenLabs, Google
• Italian: ElevenLabs, Google
• Portuguese: ElevenLabs, Google
• Polish: ElevenLabs, Google
• Turkish: ElevenLabs, Google
• Russian: ElevenLabs, Google
• Dutch: ElevenLabs, Google
• Japanese: ElevenLabs, Google
• Chinese: ElevenLabs, Google
• Korean: ElevenLabs, Google
• Hindi: ElevenLabs, Google

​

Speech-to-Text

• English, Spanish, French, German, Italian
• Portuguese, Russian, Japanese, Korean, Chinese
• Arabic, Hindi, Turkish, Polish, Dutch
• And many more…

​

Audio Formats

​

Supported Input Formats

• WAV: Uncompressed audio
• MP3: Compressed audio
• FLAC: Lossless compression
• M4A: Apple audio format
• OGG: Open source format
• WEBM: Web audio format

​

Output Formats

• MP3: Most compatible
• WAV: Uncompressed quality
• FLAC: Lossless compression
• OGG: Open source format

​

Quality Settings

• Sample Rate: 8kHz, 16kHz, 22kHz, 44.1kHz, 48kHz
• Bit Depth: 16-bit, 24-bit
• Channels: Mono, Stereo

​

Rate Limits

​

Common Use Cases

​

Best Practices

​

Text-to-Speech Optimization

• Use punctuation for natural pauses
• Spell out numbers and abbreviations
• Add SSML tags for emphasis
• Choose appropriate voice for content tone

​

Speech-to-Text Optimization

• Use high-quality audio (16kHz+)
• Minimize background noise
• Specify language for better accuracy
• Use appropriate model for use case

​

Performance Tips

• Batch process multiple files
• Use streaming for real-time applications
• Cache frequently used audio
• Implement proper error handling

​

Getting Started