Hugging Face's logo

Spaces:
TTS-AGI
/
TTS-Arena-V2
Running on CPU Upgrade

App Files Community
114

[Model Addition Request] Lightning v3.1 - 44kHz High-Fidelity TTS with Ultra-Low Latency

#111
by stalwartcoder - opened
Discussion
stalwartcoder

Hi! 👋

This is Abhishek from Smallest AI. We'd like to request the addition of Lightning v3.1 to the TTS Arena V2 leaderboard.

Model Overview

Lightning v3.1 is a high-fidelity, low-latency text-to-speech model delivering natural, expressive speech at 44 kHz native sample rate. Optimized for real-time applications with voice cloning support.

Specification Value
Model Lightning v3.1
Developer Smallest AI
Native Sample Rate 44,100 Hz
Supported Sample Rates 8,000 / 16,000 / 24,000 / 44,100 Hz
Languages English, Hindi, Spanish, Tamil
Streaming HTTP, SSE, WebSocket

API Integration

  • API Endpoint:
    • Get voices: https://waves-api.smallest.ai/api/v1/lightning-v3.1/get_voices
    • Get speech: https://waves-api.smallest.ai/api/v1/lightning-v3.1/get_speech
  • Documentation: https://waves-docs.smallest.ai
  • Authentication: Bearer token

Sample Request: 

curl --request POST \
  --url 'https://waves-api.smallest.ai/api/v1/lightning-v3.1/get_speech' \
  --header 'Authorization: Bearer YOUR_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
    "text": "Hello, welcome to the TTS Arena!",
    "voice_id": "lauren",
    "sample_rate": 24000
  }' \
  --output audio.wav

Streaming Endpoints:

Endpoint Method Use Case
/api/v1/lightning-v3.1/get_speech POST Synchronous synthesis
/api/v1/lightning-v3.1/stream POST (SSE) Server-sent events streaming
/api/v1/lightning-v3.1/get_speech/stream WebSocket Real-time streaming

Available Voices

American English:

Voice ID Name Gender
sandra Sandra Female
daniel Daniel Male

Request Parameters

Parameter Type Required Default Description
text string Yes - Text to synthesize
voice_id string Yes - Voice identifier
sample_rate integer No 24000 Output sample rate (Hz)
speed float No 1.0 Speech speed (0.5-2.0)
add_wav_header boolean No true Include WAV header

Links

I can DM you the API keys to you @mrfakename or @reach-vb

Contact

We can provide a dedicated API key for Arena testing. I can DM on HF.
I am also reachable via:

Thanks!
mrfakename
 TTS AGI org

Would love to onboard your model! Please shoot me an email at me@mrfake.name or send me a DM on Discord @realmrfakename. (HF doesn’t support DMs yet)
stalwartcoder

Hi @mrfakename , thanks for the quick response, I have sent the email.
its from abhishek [at] smallest [dot] ai

stalwartcoder

@mrfakename , let me know if you need more details.
Thanks :)

Sign up or log in to comment