par-cli-tts 0.2.2__tar.gz → 0.4.0__tar.gz

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: par-cli-tts
-Version: 0.2.2
+Version: 0.4.0
 Summary: PAR CLI TTS - Command line text-to-speech tool using ElevenLabs with voice caching and name resolution
 Project-URL: Homepage, https://github.com/paulrobello/par-cli-tts
 Project-URL: Documentation, https://github.com/paulrobello/par-cli-tts/blob/main/README.md
@@ -51,7 +51,7 @@ Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Classifier: Typing :: Typed
 Requires-Python: >=3.11
 Requires-Dist: elevenlabs>=1.0.0
-Requires-Dist: kokoro-onnx>=0.4.9
+Requires-Dist: kokoro-onnx>=0.5.0
 Requires-Dist: openai>=1.100.1
 Requires-Dist: platformdirs>=4.3.0
 Requires-Dist: python-dotenv>=1.0.0
@@ -68,7 +68,7 @@ Description-Content-Type: text/markdown
 ![Arch x86-63 | ARM | AppleSilicon](https://img.shields.io/badge/arch-x86--64%20%7C%20ARM%20%7C%20AppleSilicon-blue)
 
 ![MIT License](https://img.shields.io/badge/license-MIT-green.svg)
-![Version](https://img.shields.io/badge/version-0.2.2-green.svg)
+![Version](https://img.shields.io/badge/version-0.4.0-green.svg)
 ![Development Status](https://img.shields.io/badge/status-stable-green.svg)
 
 A powerful command-line text-to-speech tool supporting multiple TTS providers (ElevenLabs, OpenAI, and Kokoro ONNX) with intelligent voice caching, name resolution, and flexible output options.
@@ -77,6 +77,18 @@ A powerful command-line text-to-speech tool supporting multiple TTS providers (E
 
 ## What's New
 
+### v0.4.0
+- **OpenAI gpt-4o-mini-tts** - New steerable TTS model with `--instructions` option
+- **7 new OpenAI voices** - ash, ballad, coral, sage, verse, marin, cedar (13 total)
+- **ElevenLabs model updated** - Changed from deprecated `eleven_monolingual_v1` to `eleven_multilingual_v2`
+- **Kokoro ONNX 0.5.0** - Updated to latest version
+
+### v0.3.0
+- Major code refactoring with better modularity and code organization
+- New utility modules for shared console, defaults, and HTTP client
+- Test suite with 46 tests for better reliability
+- Documentation synced to match current implementation
+
 ### v0.2.2
 - Updated all HTTP requests and downloaders to ignore SSL certificate errors
 - Improves compatibility with corporate proxies and development environments
@@ -404,6 +416,10 @@ par-tts "Stable voice" --stability 0.8 --similarity 0.7
 # Adjust OpenAI speech speed
 par-tts "Fast speech" --provider openai --speed 1.5
 
+# Use OpenAI with voice instructions (gpt-4o-mini-tts only)
+par-tts "Hello there!" --provider openai --instructions "Speak in a cheerful and positive tone"
+par-tts "Good morning" -P openai -i "Speak like a pirate"
+
 # Keep temp files after playback
 par-tts "Keep this" --keep-temp
 
@@ -510,6 +526,7 @@ make clear-cache     # Clear voice cache including samples
 |--------|-------|-------------|---------|
 | `--speed` | `-r` | Speech speed (0.25 to 4.0) | 1.0 |
 | `--format` | `-f` | Audio format (mp3, opus, aac, flac, wav) | mp3 |
+| `--instructions` | `-i` | Voice instructions for gpt-4o-mini-tts (e.g., "Speak cheerfully") | None |
 
 ### Kokoro ONNX Options
 
@@ -543,7 +560,12 @@ make clear-cache     # Clear voice cache including samples
 
 ### ElevenLabs
 
-- **Models**: eleven_monolingual_v1, eleven_multilingual_v1, eleven_multilingual_v2
+- **Models**:
+  - `eleven_multilingual_v2` (default) - Most lifelike, 29 languages
+  - `eleven_v3` - Most expressive, 70+ languages
+  - `eleven_flash_v2.5` - Ultra-low latency (~75ms), 32 languages
+  - `eleven_turbo_v2.5` - Balanced quality/speed, 32 languages
+  - ~~`eleven_monolingual_v1`~~ - Deprecated, will be removed
 - **Voices**: 25+ voices with different accents and styles
 - **Features**: Voice cloning, stability control, similarity boost
 - **Smart Caching**:
@@ -555,15 +577,28 @@ make clear-cache     # Clear voice cache including samples
 
 ### OpenAI
 
-- **Models**: tts-1 (optimized for speed), tts-1-hd (optimized for quality)
-- **Voices**:
+- **Models**:
+  - `gpt-4o-mini-tts` (default) - Steerable TTS with instructions
+  - `tts-1` - Optimized for speed
+  - `tts-1-hd` - Optimized for quality
+- **Voices** (13 total):
   - alloy - Neutral and balanced
+  - ash - Enthusiastic and energetic
+  - ballad - Warm and soulful
+  - coral - Friendly and approachable
   - echo - Smooth and articulate
   - fable - Expressive and animated
-  - onyx - Deep and authoritative
   - nova - Warm and friendly (default)
+  - onyx - Deep and authoritative
+  - sage - Calm and wise
   - shimmer - Soft and gentle
-- **Features**: Speed control (0.25x to 4x), multiple output formats
+  - verse - Clear and melodic
+  - marin - Gentle and soothing
+  - cedar - Rich and resonant
+- **Features**:
+  - Speed control (0.25x to 4x)
+  - Multiple output formats
+  - Voice instructions for gpt-4o-mini-tts (steer emotion, accent, tone)
 - **Output Formats**: mp3, opus, aac, flac, wav, pcm
 - **API Key**: Set `OPENAI_API_KEY` in your .env file
 

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/README.md

@@ -5,7 +5,7 @@
 ![Arch x86-63 | ARM | AppleSilicon](https://img.shields.io/badge/arch-x86--64%20%7C%20ARM%20%7C%20AppleSilicon-blue)
 
 ![MIT License](https://img.shields.io/badge/license-MIT-green.svg)
-![Version](https://img.shields.io/badge/version-0.2.2-green.svg)
+![Version](https://img.shields.io/badge/version-0.4.0-green.svg)
 ![Development Status](https://img.shields.io/badge/status-stable-green.svg)
 
 A powerful command-line text-to-speech tool supporting multiple TTS providers (ElevenLabs, OpenAI, and Kokoro ONNX) with intelligent voice caching, name resolution, and flexible output options.
@@ -14,6 +14,18 @@ A powerful command-line text-to-speech tool supporting multiple TTS providers (E
 
 ## What's New
 
+### v0.4.0
+- **OpenAI gpt-4o-mini-tts** - New steerable TTS model with `--instructions` option
+- **7 new OpenAI voices** - ash, ballad, coral, sage, verse, marin, cedar (13 total)
+- **ElevenLabs model updated** - Changed from deprecated `eleven_monolingual_v1` to `eleven_multilingual_v2`
+- **Kokoro ONNX 0.5.0** - Updated to latest version
+
+### v0.3.0
+- Major code refactoring with better modularity and code organization
+- New utility modules for shared console, defaults, and HTTP client
+- Test suite with 46 tests for better reliability
+- Documentation synced to match current implementation
+
 ### v0.2.2
 - Updated all HTTP requests and downloaders to ignore SSL certificate errors
 - Improves compatibility with corporate proxies and development environments
@@ -341,6 +353,10 @@ par-tts "Stable voice" --stability 0.8 --similarity 0.7
 # Adjust OpenAI speech speed
 par-tts "Fast speech" --provider openai --speed 1.5
 
+# Use OpenAI with voice instructions (gpt-4o-mini-tts only)
+par-tts "Hello there!" --provider openai --instructions "Speak in a cheerful and positive tone"
+par-tts "Good morning" -P openai -i "Speak like a pirate"
+
 # Keep temp files after playback
 par-tts "Keep this" --keep-temp
 
@@ -447,6 +463,7 @@ make clear-cache     # Clear voice cache including samples
 |--------|-------|-------------|---------|
 | `--speed` | `-r` | Speech speed (0.25 to 4.0) | 1.0 |
 | `--format` | `-f` | Audio format (mp3, opus, aac, flac, wav) | mp3 |
+| `--instructions` | `-i` | Voice instructions for gpt-4o-mini-tts (e.g., "Speak cheerfully") | None |
 
 ### Kokoro ONNX Options
 
@@ -480,7 +497,12 @@ make clear-cache     # Clear voice cache including samples
 
 ### ElevenLabs
 
-- **Models**: eleven_monolingual_v1, eleven_multilingual_v1, eleven_multilingual_v2
+- **Models**:
+  - `eleven_multilingual_v2` (default) - Most lifelike, 29 languages
+  - `eleven_v3` - Most expressive, 70+ languages
+  - `eleven_flash_v2.5` - Ultra-low latency (~75ms), 32 languages
+  - `eleven_turbo_v2.5` - Balanced quality/speed, 32 languages
+  - ~~`eleven_monolingual_v1`~~ - Deprecated, will be removed
 - **Voices**: 25+ voices with different accents and styles
 - **Features**: Voice cloning, stability control, similarity boost
 - **Smart Caching**:
@@ -492,15 +514,28 @@ make clear-cache     # Clear voice cache including samples
 
 ### OpenAI
 
-- **Models**: tts-1 (optimized for speed), tts-1-hd (optimized for quality)
-- **Voices**:
+- **Models**:
+  - `gpt-4o-mini-tts` (default) - Steerable TTS with instructions
+  - `tts-1` - Optimized for speed
+  - `tts-1-hd` - Optimized for quality
+- **Voices** (13 total):
   - alloy - Neutral and balanced
+  - ash - Enthusiastic and energetic
+  - ballad - Warm and soulful
+  - coral - Friendly and approachable
   - echo - Smooth and articulate
   - fable - Expressive and animated
-  - onyx - Deep and authoritative
   - nova - Warm and friendly (default)
+  - onyx - Deep and authoritative
+  - sage - Calm and wise
   - shimmer - Soft and gentle
-- **Features**: Speed control (0.25x to 4x), multiple output formats
+  - verse - Clear and melodic
+  - marin - Gentle and soothing
+  - cedar - Rich and resonant
+- **Features**:
+  - Speed control (0.25x to 4x)
+  - Multiple output formats
+  - Voice instructions for gpt-4o-mini-tts (steer emotion, accent, tone)
 - **Output Formats**: mp3, opus, aac, flac, wav, pcm
 - **API Key**: Set `OPENAI_API_KEY` in your .env file
 

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/pyproject.toml

@@ -45,7 +45,7 @@ dependencies = [
     "platformdirs>=4.3.0",
     "pyyaml>=6.0.0",
     "openai>=1.100.1",
-    "kokoro-onnx>=0.4.9",
+    "kokoro-onnx>=0.5.0",
     "soundfile>=0.13.1",
 ]
 
@@ -78,8 +78,14 @@ dev = [
     "pre-commit>=4.2.0",
     "ruff>=0.12.7",
     "hatchling>=1.27.0",
+    "pytest>=8.0.0",
+    "pytest-cov>=5.0.0",
 ]
 
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+addopts = "-v --cov=src --cov-report=term-missing"
+
 [tool.hatch.version]
 path = "src/__init__.py"
 

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/src/__init__.py

@@ -1,3 +1,3 @@
 """PAR CLI TTS - Command line text-to-speech tool."""
 
-__version__ = "0.2.2"
+__version__ = "0.4.0"

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/src/config_file.py

@@ -6,9 +6,8 @@ from typing import Any
 import platformdirs
 import yaml
 from pydantic import BaseModel, Field, ValidationError
-from rich.console import Console
 
-console = Console()
+from src.console import console
 
 
 class ConfigFile(BaseModel):

par_cli_tts-0.4.0/src/console.py

@@ -0,0 +1,13 @@
+"""Shared console instance for consistent output handling.
+
+This module provides a single Console instance that can be imported
+by all modules for consistent output formatting.
+"""
+
+from rich.console import Console
+
+# Shared console instance for standard output
+console = Console()
+
+# Console instance for error output (writes to stderr)
+error_console = Console(stderr=True)

par_cli_tts-0.4.0/src/defaults.py

@@ -0,0 +1,33 @@
+"""Central default values for PAR CLI TTS.
+
+This module provides default configuration values for all providers
+to avoid duplication and ensure consistency.
+"""
+
+import os
+
+# Default provider
+DEFAULT_PROVIDER = "kokoro-onnx"
+
+# Default voices by provider
+DEFAULT_ELEVENLABS_VOICE = "Juniper"
+DEFAULT_OPENAI_VOICE = "nova"
+DEFAULT_KOKORO_VOICE = "af_sarah"
+
+
+def get_default_voice(provider: str) -> str:
+    """Get default voice for the specified provider.
+
+    Args:
+        provider: Provider name (elevenlabs, openai, kokoro-onnx).
+
+    Returns:
+        Default voice ID for the provider, checking environment
+        variables first, then falling back to built-in defaults.
+    """
+    defaults = {
+        "elevenlabs": os.getenv("ELEVENLABS_VOICE_ID", DEFAULT_ELEVENLABS_VOICE),
+        "openai": os.getenv("OPENAI_VOICE_ID", DEFAULT_OPENAI_VOICE),
+        "kokoro-onnx": os.getenv("KOKORO_VOICE_ID", DEFAULT_KOKORO_VOICE),
+    }
+    return defaults.get(provider, "")

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/src/errors.py

@@ -4,9 +4,7 @@ import sys
 from enum import Enum
 from typing import NoReturn
 
-from rich.console import Console
-
-console = Console(stderr=True)
+from src.console import error_console as console
 
 
 class ErrorType(Enum):

par_cli_tts-0.4.0/src/http_client.py

@@ -0,0 +1,21 @@
+"""HTTP client factory for PAR CLI TTS.
+
+This module provides a factory function for creating HTTP clients
+with consistent configuration across all providers.
+"""
+
+import httpx
+
+
+def create_http_client(timeout: float = 10.0, verify: bool = False) -> httpx.Client:
+    """Create an HTTP client with standard configuration.
+
+    Args:
+        timeout: Request timeout in seconds.
+        verify: Whether to verify SSL certificates. Defaults to False
+            to handle environments with SSL certificate issues.
+
+    Returns:
+        Configured httpx.Client instance.
+    """
+    return httpx.Client(verify=verify, timeout=timeout)

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/src/kokoro_cli.py

@@ -5,12 +5,11 @@ import sys
 from typing import Annotated
 
 import typer
-from rich.console import Console
 from rich.table import Table
 
+from src.console import console
 from src.model_downloader import ModelDownloader
 
-console: Console = Console()
 app = typer.Typer(help="Manage Kokoro ONNX TTS models")
 
 

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/src/model_downloader.py

@@ -6,10 +6,9 @@ import urllib.request
 from pathlib import Path
 
 from platformdirs import user_data_dir
-from rich.console import Console
 from rich.progress import BarColumn, DownloadColumn, Progress, SpinnerColumn, TextColumn, TransferSpeedColumn
 
-console = Console()
+from src.console import console
 
 
 class ModelDownloader:

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/src/providers/elevenlabs.py

@@ -1,19 +1,20 @@
 """ElevenLabs TTS provider implementation."""
 
+import warnings
 from collections.abc import Iterator
 from pathlib import Path
 from typing import Any
 
-import httpx
-from elevenlabs import VoiceSettings, play, save
-from elevenlabs.client import ElevenLabs
-from rich.console import Console
+with warnings.catch_warnings():
+    warnings.filterwarnings("ignore", message="Core Pydantic V1 functionality", category=UserWarning)
+    from elevenlabs import VoiceSettings, play, save
+    from elevenlabs.client import ElevenLabs
 
+from src.defaults import DEFAULT_ELEVENLABS_VOICE
+from src.http_client import create_http_client
 from src.providers.base import TTSProvider, Voice
 from src.voice_cache import VoiceCache, resolve_voice_identifier
 
-console = Console()
-
 
 class ElevenLabsProvider(TTSProvider):
     """ElevenLabs TTS provider."""
@@ -27,8 +28,8 @@ class ElevenLabsProvider(TTSProvider):
             **kwargs: Additional configuration.
         """
         super().__init__(api_key, **kwargs)
-        # Create httpx client with SSL verification disabled
-        http_client = httpx.Client(verify=False, timeout=kwargs.get("timeout", 10.0))
+        # Create httpx client with standard configuration
+        http_client = create_http_client(timeout=kwargs.get("timeout", 10.0))
         self.client = ElevenLabs(api_key=api_key, httpx_client=http_client)
         self.cache = VoiceCache(app_name="par-tts-elevenlabs")
 
@@ -45,12 +46,12 @@ class ElevenLabsProvider(TTSProvider):
     @property
     def default_model(self) -> str:
         """Default model for this provider."""
-        return "eleven_monolingual_v1"
+        return "eleven_multilingual_v2"
 
     @property
     def default_voice(self) -> str:
         """Default voice for this provider."""
-        return "aMSt68OGf4xUZAnLpTU8"
+        return DEFAULT_ELEVENLABS_VOICE
 
     def generate_speech(
         self,

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/src/providers/kokoro_onnx.py

@@ -8,7 +8,9 @@ from pathlib import Path
 import soundfile as sf
 from kokoro_onnx import Kokoro
 
+from src.defaults import DEFAULT_KOKORO_VOICE
 from src.model_downloader import ModelDownloader
+from src.utils import play_audio_bytes
 
 from .base import TTSProvider, Voice
 
@@ -76,7 +78,7 @@ class KokoroONNXProvider(TTSProvider):
     @property
     def default_voice(self) -> str:
         """Default voice ID."""
-        return "af_sarah"
+        return DEFAULT_KOKORO_VOICE
 
     def list_voices(self) -> list[Voice]:
         """List available voices.
@@ -187,42 +189,4 @@ class KokoroONNXProvider(TTSProvider):
         if not isinstance(audio_data, bytes):
             audio_data = b"".join(audio_data)
 
-        # Save to temp file and play
-        with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as temp_file:
-            temp_file.write(audio_data)
-            temp_path = temp_file.name
-
-        try:
-            # Use system command to play audio
-            import platform
-            import subprocess
-
-            system = platform.system()
-            if system == "Darwin":  # macOS
-                # afplay supports volume flag (-v)
-                subprocess.run(["afplay", "-v", str(volume), temp_path], check=True)
-            elif system == "Windows":
-                # Windows doesn't have native volume control for start command
-                subprocess.run(["start", "", temp_path], shell=True, check=True)
-            else:  # Linux and others
-                # Try different players with volume support
-                players_with_volume = [
-                    ("paplay", ["--volume", str(int(volume * 65536))]),  # paplay uses 0-65536
-                    ("ffplay", ["-volume", str(int(volume * 100)), "-nodisp", "-autoexit"]),  # ffplay uses 0-100
-                    ("mpg123", ["-f", str(int(volume * 32768))]),  # mpg123 uses scale factor
-                    ("aplay", []),  # aplay doesn't support volume directly
-                ]
-
-                for player, volume_args in players_with_volume:
-                    try:
-                        cmd = [player] + volume_args + [temp_path]
-                        subprocess.run(cmd, check=True)
-                        break
-                    except (subprocess.CalledProcessError, FileNotFoundError):
-                        continue
-                else:
-                    raise RuntimeError("No audio player found. Install aplay, paplay, ffplay, or mpg123.")
-        finally:
-            # Clean up temp file
-            if Path(temp_path).exists():
-                Path(temp_path).unlink()
+        play_audio_bytes(audio_data, volume=volume, suffix=".wav")

{par_cli_tts-0.2.2 → par_cli_tts-0.4.0}/src/providers/openai.py

@@ -1,30 +1,38 @@
 """OpenAI TTS provider implementation."""
 
-import tempfile
 from collections.abc import Iterator
 from pathlib import Path
 from typing import Any, Literal
 
-import httpx
 from openai import OpenAI
-from rich.console import Console
 
+from src.console import console
+from src.defaults import DEFAULT_OPENAI_VOICE
+from src.http_client import create_http_client
 from src.providers.base import TTSProvider, Voice
-
-console = Console()
+from src.utils import play_audio_bytes
 
 
 class OpenAIProvider(TTSProvider):
     """OpenAI TTS provider."""
 
     # Available voices for OpenAI TTS
+    # tts-1/tts-1-hd support: alloy, ash, coral, echo, fable, onyx, nova, sage, shimmer
+    # gpt-4o-mini-tts adds: ballad, verse, marin, cedar
     VOICES = {
         "alloy": "Alloy - Neutral and balanced",
+        "ash": "Ash - Enthusiastic and energetic",
+        "ballad": "Ballad - Warm and soulful",
+        "coral": "Coral - Friendly and approachable",
         "echo": "Echo - Smooth and articulate",
         "fable": "Fable - Expressive and animated",
-        "onyx": "Onyx - Deep and authoritative",
         "nova": "Nova - Warm and friendly",
+        "onyx": "Onyx - Deep and authoritative",
+        "sage": "Sage - Calm and wise",
         "shimmer": "Shimmer - Soft and gentle",
+        "verse": "Verse - Clear and melodic",
+        "marin": "Marin - Gentle and soothing",
+        "cedar": "Cedar - Rich and resonant",
     }
 
     def __init__(self, api_key: str, **kwargs: Any):
@@ -36,8 +44,8 @@ class OpenAIProvider(TTSProvider):
             **kwargs: Additional configuration.
         """
         super().__init__(api_key, **kwargs)
-        # Create httpx client with SSL verification disabled
-        http_client = httpx.Client(verify=False, timeout=kwargs.get("timeout", 10.0))
+        # Create httpx client with standard configuration
+        http_client = create_http_client(timeout=kwargs.get("timeout", 10.0))
         self.client = OpenAI(api_key=api_key, http_client=http_client)
 
     @property
@@ -53,12 +61,12 @@ class OpenAIProvider(TTSProvider):
     @property
     def default_model(self) -> str:
         """Default model for this provider."""
-        return "tts-1"
+        return "gpt-4o-mini-tts"
 
     @property
     def default_voice(self) -> str:
         """Default voice for this provider."""
-        return "nova"
+        return DEFAULT_OPENAI_VOICE
 
     def generate_speech(
         self,
@@ -67,6 +75,7 @@ class OpenAIProvider(TTSProvider):
         model: str | None = None,
         response_format: Literal["mp3", "opus", "aac", "flac", "wav", "pcm"] = "mp3",
         speed: float = 1.0,
+        instructions: str | None = None,
         **kwargs: Any,
     ) -> bytes:
         """
@@ -74,10 +83,11 @@ class OpenAIProvider(TTSProvider):
 
         Args:
             text: Text to convert to speech.
-            voice: Voice to use (alloy, echo, fable, onyx, nova, shimmer).
-            model: Model to use (tts-1 or tts-1-hd).
+            voice: Voice to use (alloy, ash, ballad, coral, echo, fable, nova, onyx, sage, shimmer, verse, marin, cedar).
+            model: Model to use (gpt-4o-mini-tts, tts-1, or tts-1-hd).
             response_format: Audio format (mp3, opus, aac, flac, wav).
             speed: Speed of speech (0.25 to 4.0).
+            instructions: Instructions for voice style (gpt-4o-mini-tts only).
             **kwargs: Additional parameters.
 
         Returns:
@@ -89,13 +99,20 @@ class OpenAIProvider(TTSProvider):
         # Ensure speed is within valid range
         speed = max(0.25, min(4.0, speed))
 
-        response = self.client.audio.speech.create(
-            model=model,
-            voice=voice,  # type: ignore
-            input=text,
-            response_format=response_format,  # type: ignore
-            speed=speed,
-        )
+        # Build request parameters
+        request_params: dict[str, Any] = {
+            "model": model,
+            "voice": voice,
+            "input": text,
+            "response_format": response_format,
+            "speed": speed,
+        }
+
+        # Add instructions parameter for gpt-4o-mini-tts model
+        if instructions and model == "gpt-4o-mini-tts":
+            request_params["instructions"] = instructions
+
+        response = self.client.audio.speech.create(**request_params)
 
         # Get audio data as bytes
         audio_bytes = response.content
@@ -177,45 +194,8 @@ class OpenAIProvider(TTSProvider):
             audio_data: Audio data to play (bytes or iterator).
             volume: Volume level (0.0 = silent, 1.0 = normal, 2.0 = double volume).
         """
-        # OpenAI doesn't provide a play function, so we'll use a temporary file
-        # and system audio player
-        import subprocess
-        import sys
-
         # Convert iterator to bytes if needed
         if not isinstance(audio_data, bytes):
             audio_data = b"".join(audio_data)
 
-        with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as tmp:
-            tmp.write(audio_data)
-            tmp_path = tmp.name
-
-        try:
-            # Determine the appropriate command based on the platform
-            if sys.platform == "darwin":  # macOS
-                # afplay supports volume flag (-v)
-                subprocess.run(["afplay", "-v", str(volume), tmp_path], check=True)
-            elif sys.platform == "win32":  # Windows
-                # Windows doesn't have native volume control for start command
-                subprocess.run(["start", "", tmp_path], shell=True, check=True)
-            else:  # Linux and others
-                # Try common audio players with volume support
-                players_with_volume = [
-                    ("paplay", ["--volume", str(int(volume * 65536))]),  # paplay uses 0-65536
-                    ("ffplay", ["-volume", str(int(volume * 100)), "-nodisp", "-autoexit"]),  # ffplay uses 0-100
-                    ("mpg123", ["-f", str(int(volume * 32768))]),  # mpg123 uses scale factor
-                    ("aplay", []),  # aplay doesn't support volume directly
-                ]
-
-                for player, volume_args in players_with_volume:
-                    try:
-                        cmd = [player] + volume_args + [tmp_path]
-                        subprocess.run(cmd, check=True)
-                        break
-                    except (subprocess.CalledProcessError, FileNotFoundError):
-                        continue
-                else:
-                    console.print("[yellow]Warning: Could not find audio player. Audio saved but not played.[/yellow]")
-        finally:
-            # Clean up temp file
-            Path(tmp_path).unlink(missing_ok=True)
+        play_audio_bytes(audio_data, volume=volume, suffix=".mp3")