ggerganov commited on
Commit
d161cee
·
unverified ·
1 Parent(s): 81dca98

Update README.md

Browse files
Files changed (1) hide show
  1. examples/talk/README.md +8 -0
examples/talk/README.md CHANGED
@@ -24,6 +24,8 @@ make talk
24
  ./talk -p Santa
25
  ```
26
 
 
 
27
  To run this, you will need a ggml GPT-2 model: [instructions](https://github.com/ggerganov/ggml/tree/master/examples/gpt-2#downloading-and-converting-the-original-models)
28
 
29
  Alternatively, you can simply download the smallest ggml GPT-2 117M model (240 MB) like this:
@@ -31,3 +33,9 @@ Alternatively, you can simply download the smallest ggml GPT-2 117M model (240 M
31
  ```
32
  wget --quiet --show-progress -O models/ggml-gpt-2-117M.bin https://ggml.ggerganov.com/ggml-model-gpt-2-117M.bin
33
  ```
 
 
 
 
 
 
 
24
  ./talk -p Santa
25
  ```
26
 
27
+ ## GPT-2
28
+
29
  To run this, you will need a ggml GPT-2 model: [instructions](https://github.com/ggerganov/ggml/tree/master/examples/gpt-2#downloading-and-converting-the-original-models)
30
 
31
  Alternatively, you can simply download the smallest ggml GPT-2 117M model (240 MB) like this:
 
33
  ```
34
  wget --quiet --show-progress -O models/ggml-gpt-2-117M.bin https://ggml.ggerganov.com/ggml-model-gpt-2-117M.bin
35
  ```
36
+
37
+ ## TTS
38
+
39
+ For best experience, this example needs a TTS tool to convert the generated text responses to voice.
40
+ You can use any TTS engine that you would like - simply edit the [speak.sh](speak.sh) script to your needs.
41
+ By default, it is configured to use `espeak`, but you can use whatever you wish.