Skip to content

SOTA Open Source TTS

Fix document (use medium by default)

Lengyue9677932a21fd
diff --git a/docs/en/finetune.md b/docs/en/finetune.md
index 694c39b..72db128 100644
--- a/docs/en/finetune.md
+++ b/docs/en/finetune.md
@@ -154,7 +154,7 @@ huggingface-cli download fishaudio/fish-speech-1 text2semantic-sft-medium-v1-4k.
Finally, you can start the fine-tuning by running the following command:
```bash
python fish_speech/train.py --config-name text2semantic_finetune \
- [email protected]=dual_ar_2_codebook_large
+ [email protected]=dual_ar_2_codebook_medium
```
!!! note
@@ -180,7 +180,7 @@ After training, you need to convert the LoRA weights to regular weights before p
```bash
python tools/llama/merge_lora.py \
- --llama-config dual_ar_2_codebook_large \
+ --llama-config dual_ar_2_codebook_medium \
--lora-config r_8_alpha_16 \
--llama-weight checkpoints/text2semantic-sft-medium-v1-4k.pth \
--lora-weight results/text2semantic-finetune-medium-lora/checkpoints/step_000000200.ckpt \
diff --git a/docs/en/inference.md b/docs/en/inference.md
index 4befc16..32973b0 100644
--- a/docs/en/inference.md
+++ b/docs/en/inference.md
@@ -37,7 +37,7 @@ python tools/llama/generate.py \
--text "The text you want to convert" \
--prompt-text "Your reference text" \
--prompt-tokens "fake.npy" \
- --config-name dual_ar_2_codebook_large \
+ --config-name dual_ar_2_codebook_medium \
--checkpoint-path "checkpoints/text2semantic-sft-medium-v1-4k.pth" \
--num-samples 2 \
--compile
@@ -70,7 +70,7 @@ We provide a HTTP API for inference. You can use the following command to start
python -m tools.api \
--listen 0.0.0.0:8000 \
--llama-checkpoint-path "checkpoints/text2semantic-sft-medium-v1-4k.pth" \
- --llama-config-name dual_ar_2_codebook_large \
+ --llama-config-name dual_ar_2_codebook_medium \
--vqgan-checkpoint-path "checkpoints/vq-gan-group-fsq-2x1024.pth"
```
@@ -83,7 +83,7 @@ You can start the WebUI using the following command:
```bash
python -m tools.webui \
--llama-checkpoint-path "checkpoints/text2semantic-sft-medium-v1-4k.pth" \
- --llama-config-name dual_ar_2_codebook_large \
+ --llama-config-name dual_ar_2_codebook_medium \
--vqgan-checkpoint-path "checkpoints/vq-gan-group-fsq-2x1024.pth"
```
diff --git a/docs/zh/finetune.md b/docs/zh/finetune.md
index d4fde2f..213fa83 100644
--- a/docs/zh/finetune.md
+++ b/docs/zh/finetune.md
@@ -165,7 +165,7 @@ HF_ENDPOINT=https://hf-mirror.com huggingface-cli download fishaudio/fish-speech
```bash
python fish_speech/train.py --config-name text2semantic_finetune \
- [email protected]=dual_ar_2_codebook_large
+ [email protected]=dual_ar_2_codebook_medium
```
!!! note
@@ -190,7 +190,7 @@ python fish_speech/train.py --config-name text2semantic_finetune \
```bash
python tools/llama/merge_lora.py \
- --llama-config dual_ar_2_codebook_large \
+ --llama-config dual_ar_2_codebook_medium \
--lora-config r_8_alpha_16 \
--llama-weight checkpoints/text2semantic-sft-medium-v1-4k.pth \
--lora-weight results/text2semantic-finetune-medium-lora/checkpoints/step_000000200.ckpt \
diff --git a/docs/zh/inference.md b/docs/zh/inference.md
index 232088c..3b67142 100644
--- a/docs/zh/inference.md
+++ b/docs/zh/inference.md
@@ -42,7 +42,7 @@ python tools/llama/generate.py \
--text "要转换的文本" \
--prompt-text "你的参考文本" \
--prompt-tokens "fake.npy" \
- --config-name dual_ar_2_codebook_large \
+ --config-name dual_ar_2_codebook_medium \
--checkpoint-path "checkpoints/text2semantic-sft-medium-v1-4k.pth" \
--num-samples 2 \
--compile
@@ -75,7 +75,7 @@ python tools/vqgan/inference.py \
python -m tools.api \
--listen 0.0.0.0:8000 \
--llama-checkpoint-path "checkpoints/text2semantic-sft-medium-v1-4k.pth" \
- --llama-config-name dual_ar_2_codebook_large \
+ --llama-config-name dual_ar_2_codebook_medium \
--vqgan-checkpoint-path "checkpoints/vq-gan-group-fsq-2x1024.pth"
# 推荐中国大陆用户运行以下命令来启动 HTTP 服务:
@@ -91,7 +91,7 @@ HF_ENDPOINT=https://hf-mirror.com python -m ...
```bash
python -m tools.webui \
--llama-checkpoint-path "checkpoints/text2semantic-sft-medium-v1-4k.pth" \
- --llama-config-name dual_ar_2_codebook_large \
+ --llama-config-name dual_ar_2_codebook_medium \
--vqgan-checkpoint-path "checkpoints/vq-gan-group-fsq-2x1024.pth"
```
diff --git a/tools/api.py b/tools/api.py
index ff900b3..670c9bc 100644
--- a/tools/api.py
+++ b/tools/api.py
@@ -227,7 +227,7 @@ def parse_args():
default="checkpoints/text2semantic-sft-medium-v1-4k.pth",
)
parser.add_argument(
- "--llama-config-name", type=str, default="dual_ar_2_codebook_large"
+ "--llama-config-name", type=str, default="dual_ar_2_codebook_medium"
)
parser.add_argument(
"--vqgan-checkpoint-path",
diff --git a/tools/llama/merge_lora.py b/tools/llama/merge_lora.py
index d125d12..5f9c2d0 100644
--- a/tools/llama/merge_lora.py
+++ b/tools/llama/merge_lora.py
@@ -12,7 +12,7 @@ from fish_speech.models.text2semantic.lora_utils import (
@click.command()
[email protected]("--llama-config", type=str, default="dual_ar_2_codebook_large")
[email protected]("--llama-config", type=str, default="dual_ar_2_codebook_medium")
@click.option("--lora-config", type=str, default="r_8_alpha_16")
@click.option(
"--llama-weight", type=str, default="checkpoints/text2semantic-sft-medium-v1-4k.pth"
diff --git a/tools/webui.py b/tools/webui.py
index 430fd1f..6186063 100644
--- a/tools/webui.py
+++ b/tools/webui.py
@@ -324,7 +324,7 @@ def parse_args():
default="checkpoints/text2semantic-sft-medium-v1-4k.pth",
)
parser.add_argument(
- "--llama-config-name", type=str, default="dual_ar_2_codebook_large"
+ "--llama-config-name", type=str, default="dual_ar_2_codebook_medium"
)
parser.add_argument(
"--vqgan-checkpoint-path",