diff --git a/Dockerfile b/Dockerfile
index e1e121c1..c567eb4b 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -21,9 +21,7 @@ WORKDIR /app
 RUN wget https://github.com/jpreprocess/jpreprocess/releases/download/v0.6.1/naist-jdic-jpreprocess.tar.gz \
     && tar xzf naist-jdic-jpreprocess.tar.gz \
     && rm naist-jdic-jpreprocess.tar.gz
-RUN wget http://downloads.sourceforge.net/open-jtalk/hts_voice_nitech_jp_atr503_m001-1.05.tar.gz \
-    && tar xzf hts_voice_nitech_jp_atr503_m001-1.05.tar.gz \
-    && rm hts_voice_nitech_jp_atr503_m001-1.05.tar.gz
+RUN git clone --depth 1 https://github.com/icn-lab/htsvoice-tohoku-f01.git
 
 FROM gcr.io/distroless/nodejs20-debian12:nonroot@sha256:015be521134f97b5f2b4c1543615eb4be907fadc8c6a52e60fd0c18f7cda0337 AS runner
 WORKDIR /app
diff --git a/README.md b/README.md
index 0b62a124..fc315544 100644
--- a/README.md
+++ b/README.md
@@ -18,15 +18,14 @@ docker run --rm -d --env-file .env ghcr.io/discordjs-japan/om:latest
 
 For copyright of dependent packages, please see package.json.
 
-### [HTS Voice "NIT ATR503 M001" version 1.05](http://downloads.sourceforge.net/open-jtalk/hts_voice_nitech_jp_atr503_m001-1.05.tar.gz)
+### [HTS voice tohoku-f01](https://github.com/icn-lab/htsvoice-tohoku-f01)
 
 > [!NOTE]
 > HTS Voice is only included in Docker container
 
-Creative Commons Attribution 3.0
+Creative Commons Attributions 4.0
 
-- Copyright (c) 2003-2012  Nagoya Institute of Technology Department of Computer Science
-- Copyright (c) 2003-2008  Tokyo Institute of Technology Interdisciplinary Graduate School of Science and Engineering
+- Copyright (c) 2015 Intelligent Communication Network (Ito-Nose) Laboratory, Tohoku University.
 
 ### [naist-jdic](https://github.com/jpreprocess/jpreprocess/releases/download/v0.6.1/naist-jdic-jpreprocess.tar.gz)
 
diff --git a/src/synthesis/index.ts b/src/synthesis/index.ts
index d442ffbe..53e430d6 100644
--- a/src/synthesis/index.ts
+++ b/src/synthesis/index.ts
@@ -4,5 +4,5 @@ import WorkerSynthesizer from "./worker-synthesizer";
 export const synthesizer: Synthesizer = new WorkerSynthesizer(
   process.env.DICTIONARY ?? "/app/model/naist-jdic",
   process.env.MODEL ??
-    "/app/model/hts_voice_nitech_jp_atr503_m001-1.05/nitech_jp_atr503_m001.htsvoice",
+    "/app/model/htsvoice-tohoku-f01/tohoku-f01-neutral.htsvoice",
 );
diff --git a/src/synthesis/options.ts b/src/synthesis/options.ts
index 943e0f34..f04b209e 100644
--- a/src/synthesis/options.ts
+++ b/src/synthesis/options.ts
@@ -2,5 +2,8 @@ import { Message } from "discord.js";
 import { SynthesisOption } from "node-altjtalk-binding";
 
 export function createSynthesisOption(_: Message): SynthesisOption {
-  return {};
+  return {
+    additionalHalfTone: -5,
+    speechSpeedRate: 1.2,
+  };
 }