Update app.py
Browse files
app.py
CHANGED
|
@@ -78,7 +78,7 @@ def transcribe(audio, text):
|
|
| 78 |
buffer = []
|
| 79 |
for sentence in sentences:
|
| 80 |
sentence_tokens = tokenizer.encode(sentence)
|
| 81 |
-
if len(buffer) + len(sentence_tokens) >
|
| 82 |
subinput_tokens.append(buffer)
|
| 83 |
buffer = []
|
| 84 |
buffer.extend(sentence_tokens)
|
|
@@ -92,7 +92,7 @@ def transcribe(audio, text):
|
|
| 92 |
messages.append({"role": "user", "content": transcript["text"]+subinput_text})
|
| 93 |
|
| 94 |
num_tokens = sum(len(tokenizer.encode(message["content"])) for message in messages)
|
| 95 |
-
if num_tokens >
|
| 96 |
# Concatenate the chat history
|
| 97 |
chat_transcript = "\n\n".join([f"[ANSWER {answer_count}]{message['role']}: {message['content']}" for message in messages if message['role'] != 'user'])
|
| 98 |
# Append the number of tokens used to the end of the chat transcript
|
|
|
|
| 78 |
buffer = []
|
| 79 |
for sentence in sentences:
|
| 80 |
sentence_tokens = tokenizer.encode(sentence)
|
| 81 |
+
if len(buffer) + len(sentence_tokens) > 400:
|
| 82 |
subinput_tokens.append(buffer)
|
| 83 |
buffer = []
|
| 84 |
buffer.extend(sentence_tokens)
|
|
|
|
| 92 |
messages.append({"role": "user", "content": transcript["text"]+subinput_text})
|
| 93 |
|
| 94 |
num_tokens = sum(len(tokenizer.encode(message["content"])) for message in messages)
|
| 95 |
+
if num_tokens > 1400:
|
| 96 |
# Concatenate the chat history
|
| 97 |
chat_transcript = "\n\n".join([f"[ANSWER {answer_count}]{message['role']}: {message['content']}" for message in messages if message['role'] != 'user'])
|
| 98 |
# Append the number of tokens used to the end of the chat transcript
|