Instructions to use reach-vb/parler-expresso-v0.2-spk-10-epc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use reach-vb/parler-expresso-v0.2-spk-10-epc with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="reach-vb/parler-expresso-v0.2-spk-10-epc")

# Load model directly
from transformers import AutoModelForSeq2SeqLM
model = AutoModelForSeq2SeqLM.from_pretrained("reach-vb/parler-expresso-v0.2-spk-10-epc", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use reach-vb/parler-expresso-v0.2-spk-10-epc with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "reach-vb/parler-expresso-v0.2-spk-10-epc"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "reach-vb/parler-expresso-v0.2-spk-10-epc",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/reach-vb/parler-expresso-v0.2-spk-10-epc

SGLang

How to use reach-vb/parler-expresso-v0.2-spk-10-epc with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "reach-vb/parler-expresso-v0.2-spk-10-epc" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "reach-vb/parler-expresso-v0.2-spk-10-epc",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "reach-vb/parler-expresso-v0.2-spk-10-epc" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "reach-vb/parler-expresso-v0.2-spk-10-epc",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use reach-vb/parler-expresso-v0.2-spk-10-epc with Docker Model Runner:
```
docker model run hf.co/reach-vb/parler-expresso-v0.2-spk-10-epc
```

reach-vb commited on May 11, 2024

Commit

69c6721

1 Parent(s): be9e287

Saving train state of step 710

Browse files

Files changed (7) hide show

checkpoint-710-epoch-9/optimizer.bin +3 -0
checkpoint-710-epoch-9/pytorch_model.bin +3 -0
checkpoint-710-epoch-9/random_states_0.pkl +3 -0
checkpoint-710-epoch-9/scheduler.bin +3 -0
config.json +1 -1
generation_config.json +12 -0
model.safetensors +3 -0

checkpoint-710-epoch-9/optimizer.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:de0dc66361c4365d81c900b4a70b1c28b74d124498b5e793f5b25a8c74b9f384
+size 3652763351

checkpoint-710-epoch-9/pytorch_model.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:8e03e91d863f7b3c75f188844100bfbb9ddd34302dd6b9e3ca297ca030323b86
+size 2588462170

checkpoint-710-epoch-9/random_states_0.pkl ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:c3773120cb7ffe8a1f43fe58ceca0964f2e57bcbfaebab97658509bddd84633b
+size 14280

checkpoint-710-epoch-9/scheduler.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:984df56afcabb7169e315d994f3b0e5ff3166076ad03273594006ac7ea472dfa
+size 1000

config.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/fsx/yoach/tmp/artefacts/training-400M-punctuated-v2/",
   "architectures": [
     "ParlerTTSForConditionalGeneration"
   ],

 {
+  "_name_or_path": "parler-tts/parler_tts_mini_v0.1",
   "architectures": [
     "ParlerTTSForConditionalGeneration"
   ],

generation_config.json ADDED Viewed

	@@ -0,0 +1,12 @@

+{
+  "_from_model_config": true,
+  "bos_token_id": 1025,
+  "decoder_start_token_id": 1025,
+  "do_sample": true,
+  "eos_token_id": 1024,
+  "guidance_scale": 1.0,
+  "max_length": 2580,
+  "min_new_tokens": 50,
+  "pad_token_id": 1024,
+  "transformers_version": "4.40.1"
+}

model.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:246af7f1ddab47a7be7e2099ca21fedc0e5590a6a631134540de1beae3db74a3
+size 2588215392