whazaa 0.6.13 → 0.6.14
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +100 -0
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -67,6 +67,106 @@ The Kokoro TTS model (~160 MB) is downloaded automatically on first use of `what
|
|
|
67
67
|
|
|
68
68
|
---
|
|
69
69
|
|
|
70
|
+
## How to Use
|
|
71
|
+
|
|
72
|
+
Once Whazaa is set up, you talk to PAI (your AI assistant) in plain language. You never need to know about tool names or parameters — just say what you want.
|
|
73
|
+
|
|
74
|
+
### Sending Messages
|
|
75
|
+
|
|
76
|
+
Tell PAI what to say and to whom:
|
|
77
|
+
|
|
78
|
+
- "Send Randolf a message saying I'll be late"
|
|
79
|
+
- "Tell Nicole the meeting is moved to 3pm"
|
|
80
|
+
- "Message my self-chat: pick up milk"
|
|
81
|
+
|
|
82
|
+
If you don't say who to send it to, PAI sends to your own WhatsApp — useful for notes to yourself.
|
|
83
|
+
|
|
84
|
+
### Voice Notes
|
|
85
|
+
|
|
86
|
+
PAI can send a WhatsApp voice note instead of a text message:
|
|
87
|
+
|
|
88
|
+
- "Send me a voice note saying good morning"
|
|
89
|
+
- "Send a voice note to Nicole saying I'm on my way"
|
|
90
|
+
- "Tell George via voice note that dinner is at 7"
|
|
91
|
+
|
|
92
|
+
You can choose whose voice to use:
|
|
93
|
+
|
|
94
|
+
- "Say it as George" or "Use George's voice"
|
|
95
|
+
- "Send that as a voice note in Daniel's voice"
|
|
96
|
+
- "Use Nicole's voice for this"
|
|
97
|
+
|
|
98
|
+
See the full voice list at the bottom of this section.
|
|
99
|
+
|
|
100
|
+
### Listening Locally (Mac Speakers)
|
|
101
|
+
|
|
102
|
+
PAI can speak out loud through your Mac — no WhatsApp needed:
|
|
103
|
+
|
|
104
|
+
- "Say that out loud"
|
|
105
|
+
- "Read that to me"
|
|
106
|
+
- "Talk to me" or "Say it through the speakers"
|
|
107
|
+
|
|
108
|
+
Great for when you want an audio response right now, without sending anything to your phone.
|
|
109
|
+
|
|
110
|
+
### Voice Mode — Hands-Free
|
|
111
|
+
|
|
112
|
+
Instead of switching to voice one message at a time, you can put PAI into a persistent voice mode so every response comes back as audio automatically.
|
|
113
|
+
|
|
114
|
+
**Voice notes to your phone:**
|
|
115
|
+
|
|
116
|
+
- "Voice mode on" or "Respond via voice" — every PAI response becomes a WhatsApp voice note
|
|
117
|
+
- "Back to text" or "Text mode" — back to normal text messages
|
|
118
|
+
|
|
119
|
+
**Audio through your Mac speakers:**
|
|
120
|
+
|
|
121
|
+
- "Talk to me locally" or "Local voice mode" — every response plays through your speakers
|
|
122
|
+
- "Back to text" — turns it off
|
|
123
|
+
|
|
124
|
+
Voice mode is perfect for driving, cooking, or any time you can't look at a screen.
|
|
125
|
+
|
|
126
|
+
### Switching Voices
|
|
127
|
+
|
|
128
|
+
PAI's default voice is Fable (British male). You can switch voices by name:
|
|
129
|
+
|
|
130
|
+
- "Hi Nicole" — switches to Nicole's voice
|
|
131
|
+
- "Hi George" — switches to George's voice
|
|
132
|
+
- "Hi Daniel" — switches to Daniel's voice
|
|
133
|
+
- "Back to PAI" or "Default voice" — back to Fable
|
|
134
|
+
|
|
135
|
+
Voice switches are remembered for the session. You can also set a different default in the config.
|
|
136
|
+
|
|
137
|
+
### Chat History
|
|
138
|
+
|
|
139
|
+
PAI can look up your WhatsApp conversations directly — it reads from WhatsApp Desktop's local database, so it's fast and doesn't require your phone to be online:
|
|
140
|
+
|
|
141
|
+
- "Show me my chats" — lists your recent conversations
|
|
142
|
+
- "Show messages from Randolf" — shows recent messages from that contact
|
|
143
|
+
- "What did Nicole say last?" — PAI finds the conversation and reads it
|
|
144
|
+
|
|
145
|
+
### Session Management (from Your Phone)
|
|
146
|
+
|
|
147
|
+
You can control your PAI sessions from WhatsApp itself. Send these commands to your self-chat:
|
|
148
|
+
|
|
149
|
+
- `/s` — see a list of your active PAI sessions (each Claude window is a session)
|
|
150
|
+
- `/2` — switch to session 2
|
|
151
|
+
- `/2 Cooking Project` — switch to session 2 and name it
|
|
152
|
+
|
|
153
|
+
This is useful when you have multiple Claude windows open for different projects.
|
|
154
|
+
|
|
155
|
+
### Available Voices
|
|
156
|
+
|
|
157
|
+
28 voices across four categories:
|
|
158
|
+
|
|
159
|
+
| Category | Voices |
|
|
160
|
+
|----------|--------|
|
|
161
|
+
| American Female | af_heart, af_alloy, af_aoede, af_bella, af_jessica, af_kore, af_nicole, af_nova, af_river, af_sarah, af_sky |
|
|
162
|
+
| American Male | am_adam, am_echo, am_eric, am_fenrir, am_liam, am_michael, am_onyx, am_puck, am_santa |
|
|
163
|
+
| British Female | bf_alice, bf_emma, bf_isabella, bf_lily |
|
|
164
|
+
| British Male | bm_daniel, **bm_fable** (PAI default), bm_george, bm_lewis |
|
|
165
|
+
|
|
166
|
+
All TTS synthesis runs locally on your Mac — no audio is ever sent to any external service.
|
|
167
|
+
|
|
168
|
+
---
|
|
169
|
+
|
|
70
170
|
## MCP tools
|
|
71
171
|
|
|
72
172
|
Once configured, Claude Code has ten tools available:
|