mayaresearch/veena-max 🔢📝❓✓ → 🖼️

▶️ 40 runs 📅 Sep 2025 ⚙️ Cog 0.16.6 ⚖️ License

bilingual-tts hindi-english hindi-tts text-to-speech

About

VeenaMAX is a high-performance, advanced Text-to-Speech API that converts text into natural-sounding speech with emotional intelligence and blazing-fast response times. It can speak English and Hindi - the most widely used languages in India. As a signifi

Example Output

Output

Performance Metrics

8.87s Prediction Time

16.75s Total Time

All Input Parameters

{
  "text": "Hi! I'm VeenaMAX, a Text-to-Speech API by Maya Research. I turn text into natural English and Hindi speech with emotional intelligence.",
  "top_p": 0.9,
  "speaker": "vinaya_assist",
  "temperature": 0.5,
  "output_format": "wav",
  "max_new_tokens": 800,
  "normalize_text": true,
  "repetition_penalty": 1.1
}

Input Parameters

seed Type: integer: Random seed for reproducible generation
text Type: stringDefault: Hi! I'm VeenaMAX, a Text-to-Speech API by Maya Research. I turn text into natural English and Hindi speech with emotional intelligence.: Text to convert to speech (Hindi/English). Minimum ~30 characters required.
top_p Type: numberDefault: 0.9Range: 0 - 1: Nucleus sampling threshold. Controls diversity of generation.
speaker Default: vinaya_assist: Voice to use for synthesis
temperature Type: numberDefault: 0.5Range: 0 - 2: Controls randomness in generation. Lower values make output more focused.
output_format Default: wav: Output audio format
max_new_tokens Type: integerDefault: 800Range: 100 - 4096: Maximum number of tokens to generate
normalize_text Type: booleanDefault: true: Apply text normalization (converts numbers to words, etc.)
repetition_penalty Type: numberDefault: 1.1Range: 1 - 2: Penalty for repeated tokens. Higher values reduce repetition.

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

Generating audio for text: 'Hi! I'm VeenaMAX, a Text-to-Speech API by Maya Research. I turn text into natural English and Hindi ...' with speaker: vinaya_assist
Audio generated successfully: /tmp/tmpm33oxvps.wav

Version Details

Version ID: ed6b8d07d5b00f97a1217bed0fee35a8e1d7553f3fd982b88252bfbd92636675
Version Created: September 9, 2025

Run on Replicate →