Source script

Converted using the script https://github.com/CrispStrobe/CrispASR/blob/main/models/convert-piper-to-gguf.py the original model is https://huggingface.co/rhasspy/piper-voices/tree/main/hi/hi_IN/rohan/medium

Pre requisites

You need to download and install ffmpeg and crispasr.

How to run the model

For Windows

Assuming the model is in models diectory and the binary is in bin directory

bin\crispasr.exe --server --backend piper -m "models\hi_IN-rohan-medium.gguf" --port 8089 -l hi -t 8 --no-spoken-disclaimer --accept-marking-responsibility

For linux

./crispasr --server --backend piper -m "models/hi_IN-rohan-medium.gguf" --port 8089 -l hi -t 8 --no-spoken-disclaimer --accept-marking-responsibility

How to test

Windows

  curl -s http://localhost:8089/v1/audio/speech ^
  -H "Content-Type: application/json" ^
  -d "{\"model\":\"piper\",\"input\":\"हेलो, आप आज कैसे हैं? आज आपने बहुत अच्छा काम किया, मुझे आप पर बहुत गर्व है।\",\"spoken_disclaimer\":false,\"stream\":true,\"response_format\":\"pcm\"}" ^
  | ffplay -f s16le -ar 22050 -nodisp -

For Linux

  curl -s http://localhost:8089/v1/audio/speech \
  -H "Content-Type: application/json" \
  -d "{\"model\":\"piper\",\"input\":\"हेलो, आप आज कैसे हैं? आज आपने बहुत अच्छा काम किया, मुझे आप पर बहुत गर्व है।\",\"spoken_disclaimer\":false,\"stream\":true,\"response_format\":\"pcm\"}" \
  | ffplay -f s16le -ar 22050 -nodisp -
Downloads last month
17
GGUF
Model size
15.7M params
Architecture
piper
Hardware compatibility
Log In to add your hardware

We're not able to determine the quantization variants.

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for pronoobie/piper-voices-hindi

Quantized
(30)
this model