Use Claude Code, Codex, VSCode, Pi, and OpenCode (and 6 other harnesses) for free (1.3B+ free tokens) from your terminal, app, IDE, or phone, and now from the browser with native browser sessions (multi-harness + multi-model) like OpenClaw (voice supported + ToS friendly)
Python
Update Whisper model configuration to use 'base' as the default model ID
Alishahryar1c35ecba9d8aa
diff --git a/.env.example b/.env.exampleindex b3b93941..74a52c72 100644--- a/.env.example+++ b/.env.example@@ -35,7 +35,7 @@ MESSAGING_RATE_WINDOW=1# Voice Note Transcription (transformers Whisper; install with: uv sync --extra voice)VOICE_NOTE_ENABLED=true# WHISPER_MODEL: Hugging Face ID or short name (tiny, base, small, medium, large-v2, large-v3, large-v3-turbo)-WHISPER_MODEL=openai/whisper-large-v3-turbo+WHISPER_MODEL=baseHF_TOKEN=""diff --git a/README.md b/README.mdindex 18bb309e..6953a064 100644--- a/README.md+++ b/README.md@@ -248,7 +248,7 @@ uv sync --extra voice| Variable | Description | Default ||----------|-------------|---------|| `VOICE_NOTE_ENABLED` | Enable voice note handling | `true` |-| `WHISPER_MODEL` | Hugging Face model ID or short name (`tiny`, `base`, `small`, `medium`, `large-v2`, `large-v3`, `large-v3-turbo`) | `openai/whisper-large-v3-turbo` |+| `WHISPER_MODEL` | Hugging Face model ID or short name (`tiny`, `base`, `small`, `medium`, `large-v2`, `large-v3`, `large-v3-turbo`) | `base` || `WHISPER_DEVICE` | `cpu` \| `cuda` | `cpu` || `HF_TOKEN` | Hugging Face token for faster model downloads (optional; [create one](https://huggingface.co/settings/tokens)) | — |diff --git a/config/settings.py b/config/settings.pyindex 09b92238..7142d7dc 100644--- a/config/settings.py+++ b/config/settings.py@@ -78,10 +78,8 @@ class Settings(BaseSettings):)# Hugging Face token for faster model downloads (optional)hf_token: str = Field(default="", validation_alias="HF_TOKEN")- # Hugging Face Whisper model ID (e.g. openai/whisper-large-v3-turbo) or short name- whisper_model: str = Field(- default="openai/whisper-large-v3-turbo", validation_alias="WHISPER_MODEL"- )+ # Hugging Face Whisper model ID (e.g. openai/whisper-base) or short name+ whisper_model: str = Field(default="base", validation_alias="WHISPER_MODEL")# Device: "cpu" | "cuda"whisper_device: str = Field(default="cpu", validation_alias="WHISPER_DEVICE")diff --git a/messaging/transcription.py b/messaging/transcription.pyindex 62f618c2..70346b66 100644--- a/messaging/transcription.py+++ b/messaging/transcription.py@@ -85,7 +85,7 @@ def transcribe_audio(file_path: Path,mime_type: str,*,- whisper_model: str = "openai/whisper-large-v3-turbo",+ whisper_model: str = "base",whisper_device: str = "cpu",) -> str:"""@@ -94,7 +94,7 @@ def transcribe_audio(Args:file_path: Path to audio file (OGG, MP3, MP4, WAV, M4A supported)mime_type: MIME type of the audio (e.g. "audio/ogg")- whisper_model: Model ID (e.g. "openai/whisper-large-v3-turbo") or short name+ whisper_model: Model ID (e.g. "openai/whisper-base") or short namewhisper_device: "cpu" | "cuda"Returns: