readium · ronibhakta1 · Jul 10, 2026 · Jul 10, 2026
diff --git a/README.md b/README.md
@@ -36,9 +36,11 @@ make dev-docker       # start server — downloads models on first run
 
 Server: `http://localhost:8000` · Interactive docs: `http://localhost:8000/docs` · Demo: `http://localhost:8000/demo`
 
+**Live demo:** [speech-server.readium.org/demo](https://speech-server.readium.org/demo)
+
 **Quick test:**
 ```bash
-curl -s -X POST http://localhost:8000/v1/synthesize \
+curl -s -X POST http://localhost:8000/synthesize \
   -H 'Content-Type: application/json' \
   -d '{"text":"Hello world","voice":"urn:readium:tts:pocket:en-alba"}' \
   -o /tmp/speech.mp3 && open /tmp/speech.mp3

diff --git a/app/api/router.py b/app/api/router.py
@@ -0,0 +1,8 @@
+from fastapi import APIRouter
+
+from app.api.routes.synthesize import router as synthesize_router
+from app.api.routes.voices import router as voices_router
+
+router = APIRouter()
+router.include_router(voices_router)
+router.include_router(synthesize_router)
diff --git a/app/api/v1/routes/health.py → app/api/routes/health.py b/app/api/v1/routes/health.py → app/api/routes/health.py
diff --git a/app/api/v1/routes/synthesize.py → app/api/routes/synthesize.py b/app/api/v1/routes/synthesize.py → app/api/routes/synthesize.py
diff --git a/app/api/v1/routes/voices.py → app/api/routes/voices.py b/app/api/v1/routes/voices.py → app/api/routes/voices.py
diff --git a/app/api/v1/router.py b/app/api/v1/router.py
diff --git a/app/config/settings.py b/app/config/settings.py
@@ -15,7 +15,6 @@ class Settings(BaseSettings):
     log_level: str = "INFO"
     host: str = "0.0.0.0"
     port: int = Field(default=8000, gt=0, le=65535)
-    api_v1_prefix: str = "/v1"
     workers: int = Field(default=1, ge=1)
     domain: str = ""
 

diff --git a/app/main.py b/app/main.py
@@ -6,8 +6,8 @@
 from starlette.middleware.trustedhost import TrustedHostMiddleware
 
 from app.api.errors import register_error_handlers
-from app.api.v1.router import v1_router
-from app.api.v1.routes.health import router as health_router
+from app.api.router import router
+from app.api.routes.health import router as health_router
 from app.config.settings import settings
 from app.core.concurrency import init_semaphore
 from app.core.registry import ProviderRegistry
@@ -73,7 +73,7 @@ def create_app() -> FastAPI:
     register_error_handlers(app)
 
     app.include_router(health_router)
-    app.include_router(v1_router, prefix=settings.api_v1_prefix)
+    app.include_router(router)
 
     @app.get("/demo", include_in_schema=False)
     async def demo() -> FileResponse:

diff --git a/app/static/demo.html b/app/static/demo.html
@@ -143,7 +143,7 @@
 <body>
 <span class="eyebrow">TTS API Demo</span>
 <h1>Readium Speech Server</h1>
-<p>Talks to <code>GET /v1/voices</code> and <code>POST /v1/synthesize</code> on this same origin.</p>
+<p>Talks to <code>GET /voices</code> and <code>POST /synthesize</code> on this same origin.</p>
 
 <div class="card">
   <label for="text">Utterance</label>
@@ -266,7 +266,7 @@ <h2>
 function updateCurl() {
   const body = JSON.stringify(requestBody());
   curlEl.textContent =
-    `curl -s -X POST ${location.origin}/v1/synthesize \\\n` +
+    `curl -s -X POST ${location.origin}/synthesize \\\n` +
     `  -H 'Content-Type: application/json' \\\n` +
     `  -d '${body}' \\\n` +
     `  -o speech.${boundaryEl.checked ? 'json' : formatEl.value}`;
@@ -315,7 +315,7 @@ <h2>
 genderFilterEl.addEventListener('change', renderVoiceOptions);
 
 async function loadVoices() {
-  const resp = await fetch('/v1/voices');
+  const resp = await fetch('/voices');
   allVoices = await resp.json();
   populateFilterOptions(langFilterEl, [...new Set(allVoices.map(v => v.language))].sort());
   const providers = [...new Set(allVoices.map(v => v.provider))].sort();
@@ -329,7 +329,7 @@ <h2>
   statusPillEl.style.display = 'none';
   outputEl.classList.remove('error');
   outputEl.textContent = 'Loading...';
-  const resp = await fetch('/v1/synthesize', {
+  const resp = await fetch('/synthesize', {
     method: 'POST',
     headers: { 'Content-Type': 'application/json' },
     body: JSON.stringify(requestBody()),

diff --git a/docs/API.md b/docs/API.md
@@ -11,8 +11,8 @@ Base path: `/`. All bodies are `application/json` unless noted.
 - [Authentication](#authentication)
 - [Errors](#errors)
 - [Health](#health)
-- [`GET /voices`](#get-v1voices)
-- [`POST /synthesize`](#post-v1synthesize)
+- [`GET /voices`](#get-voices)
+- [`POST /synthesize`](#post-synthesize)
 - [Not implemented](#not-implemented)
 
 ---
@@ -65,7 +65,7 @@ Pydantic schema errors (`422`) additionally carry an `errors` array (raw Pydanti
 
 ---
 
-## `GET /v1/voices`
+## `GET /voices`
 
 ```
 GET /voices
@@ -167,7 +167,7 @@ Only `text` and `voice` are required; everything else defaults as shown.
 | `text` | string | — | Max `MAX_TEXT_LENGTH` chars (2000 default). Rejected if empty/whitespace after trim |
 | `ssml` | bool | `false` | PocketTTS strips tags before synthesis (regex `<[^>]+>` removal) — no SSML-aware prosody |
 | `language` | string \| null | `null` | Hint only; voice resolution is by `voiceURI`, not `language` |
-| `voice` | string | — | Must exactly match a `voiceURI` from `/v1/voices`. 404 if not found |
+| `voice` | string | — | Must exactly match a `voiceURI` from `/voices`. 404 if not found |
 | `prev_utterance` / `next_utterance` | string \| null | `null` | Accepted, passed into `SynthesisParams`; PocketTTS ignores both |
 | `publication_id` | string \| null | `null` | Accepted, currently unused (reserved for future cache scoping) |
 | `boundary` | bool | `false` | `true` → JSON response with base64 audio + timing marks instead of raw binary |

diff --git a/docs/configuration.md b/docs/configuration.md
@@ -56,4 +56,4 @@ First-time setup writes `APP_ENV=production` and prompts for `DOMAIN`. `make sta
 
 For local dev, set `APP_ENV=development` in `.env` (`DOMAIN` becomes optional) and use `make dev-docker` — exposes `:8000` directly with no nginx.
 
-nginx rate-limits `/v1/synthesize` (2 req/s, burst 4) and caps connections per IP. A `503` from behind nginx is a plain nginx error page, not the app's Problem Details JSON.
+nginx rate-limits `/synthesize` (2 req/s, burst 4) and caps connections per IP. A `503` from behind nginx is a plain nginx error page, not the app's Problem Details JSON.
diff --git a/docs/development.md b/docs/development.md
@@ -43,7 +43,7 @@ No changes to routes, synthesizer, or voice catalog. Language filtering and boun
 
 ```
 Client
-  └─ POST /v1/synthesize
+  └─ POST /synthesize
        └─ Synthesizer
             ├─ validate text length + content
             ├─ resolve voiceURI → (provider, voiceURI)  (VoiceCatalog)

diff --git a/nginx/templates/20-speech-server.conf.template b/nginx/templates/20-speech-server.conf.template
@@ -55,7 +55,7 @@ server {
     # meant a normal page load (docs UI, healthz, favicon, multiple parallel
     # requests on a hard refresh) could trip nginx's built-in 503 rate-limit
     # rejection well before actually threatening the synthesis backend.
-    location = /v1/synthesize {
+    location = /synthesize {
         limit_except POST {
             deny all;
         }
@@ -73,7 +73,7 @@ server {
 
         # Headroom above what a real browser opens per origin (~6-8) — this
         # is just a coarse abuse guard for general routes, not the synthesis
-        # concurrency budget (that's the /v1/synthesize block above).
+        # concurrency budget (that's the /synthesize block above).
         limit_conn perip 20;
 
         include /etc/nginx/snippets/proxy-pass.conf;

diff --git a/scripts/configure.sh b/scripts/configure.sh
@@ -229,7 +229,6 @@ APP_ENV=production
 LOG_LEVEL=INFO
 HOST=0.0.0.0
 PORT=8000
-API_V1_PREFIX=/v1
 WORKERS=1
 DOMAIN=
 

diff --git a/tests/test_errors.py b/tests/test_errors.py
@@ -2,7 +2,7 @@
 from httpx import AsyncClient
 
 _VOICE = "urn:readium:tts:fake:en-US-standard"
-_URL = "/v1/synthesize"
+_URL = "/synthesize"
 
 
 @pytest.mark.route
@@ -21,7 +21,7 @@ async def test_text_too_long_returns_413(client: AsyncClient) -> None:
 
 @pytest.mark.route
 async def test_unknown_route_returns_about_blank_problem(client: AsyncClient) -> None:
-    resp = await client.get("/v1/does-not-exist")
+    resp = await client.get("/does-not-exist")
     assert resp.status_code == 404
     assert resp.headers["content-type"] == "application/problem+json"
     body = resp.json()

diff --git a/tests/test_formats.py b/tests/test_formats.py
@@ -2,7 +2,7 @@
 from httpx import AsyncClient
 
 _VOICE = "urn:readium:tts:fake:en-US-standard"
-_SYNTH = "/v1/synthesize"
+_SYNTH = "/synthesize"
 
 
 def _body(fmt: str) -> dict:  # type: ignore[type-arg]

diff --git a/tests/test_synthesize.py b/tests/test_synthesize.py
@@ -5,7 +5,7 @@
 
 _VOICE = "urn:readium:tts:fake:en-US-standard"
 _VOICE_FR = "urn:readium:tts:fake:fr-FR-standard"
-_URL = "/v1/synthesize"
+_URL = "/synthesize"
 _WAV = {"output": {"format": "wav"}}
 
 

diff --git a/tests/test_voices.py b/tests/test_voices.py
@@ -4,7 +4,7 @@
 
 @pytest.mark.route
 async def test_list_voices_returns_all(client: AsyncClient) -> None:
-    resp = await client.get("/v1/voices")
+    resp = await client.get("/voices")
     assert resp.status_code == 200
     voices = resp.json()
     assert len(voices) == 2
@@ -15,7 +15,7 @@ async def test_list_voices_returns_all(client: AsyncClient) -> None:
 
 @pytest.mark.route
 async def test_list_voices_shape(client: AsyncClient) -> None:
-    resp = await client.get("/v1/voices")
+    resp = await client.get("/voices")
     voice = resp.json()[0]
     required = (
         "source",
@@ -35,7 +35,7 @@ async def test_list_voices_shape(client: AsyncClient) -> None:
 
 @pytest.mark.route
 async def test_filter_by_language(client: AsyncClient) -> None:
-    resp = await client.get("/v1/voices", params={"language": "en"})
+    resp = await client.get("/voices", params={"language": "en"})
     assert resp.status_code == 200
     voices = resp.json()
     assert len(voices) == 1
@@ -44,13 +44,13 @@ async def test_filter_by_language(client: AsyncClient) -> None:
 
 @pytest.mark.route
 async def test_filter_by_provider(client: AsyncClient) -> None:
-    resp = await client.get("/v1/voices", params={"provider": "fake"})
+    resp = await client.get("/voices", params={"provider": "fake"})
     assert resp.status_code == 200
     assert len(resp.json()) == 2
 
 
 @pytest.mark.route
 async def test_filter_by_unknown_language_returns_empty(client: AsyncClient) -> None:
-    resp = await client.get("/v1/voices", params={"language": "zh"})
+    resp = await client.get("/voices", params={"language": "zh"})
     assert resp.status_code == 200
     assert resp.json() == []
Original file line number	Diff line number	Diff line change
Expand Up		@@ -56,4 +56,4 @@ First-time setup writes `APP_ENV=production` and prompts for `DOMAIN`. `make sta

		For local dev, set `APP_ENV=development` in `.env` (`DOMAIN` becomes optional) and use `make dev-docker` — exposes `:8000` directly with no nginx.

		nginx rate-limits `/v1/synthesize` (2 req/s, burst 4) and caps connections per IP. A `503` from behind nginx is a plain nginx error page, not the app's Problem Details JSON.
		nginx rate-limits `/synthesize` (2 req/s, burst 4) and caps connections per IP. A `503` from behind nginx is a plain nginx error page, not the app's Problem Details JSON.